Search

Your search keyword '"Wang Robert"' showing total 1,738 results
Start Over Author "Wang Robert"
1,738 results on '"Wang Robert"'

1. Spectral Sparsification by Deterministic Discrepancy Walk

Author

Lau, Lap Chi, Wang, Robert, and Zhou, Hong

Subjects
Computer Science - Data Structures and Algorithms, Mathematics - Combinatorics
Abstract

Spectral sparsification and discrepancy minimization are two well-studied areas that are closely related. Building on recent connections between these two areas, we generalize the "deterministic discrepancy walk" framework by Pesenti and Vladu [SODA~23] for vector discrepancy to matrix discrepancy, and use it to give a simpler proof of the matrix partial coloring theorem of Reis and Rothvoss [SODA~20]. Moreover, we show that this matrix discrepancy framework provides a unified approach for various spectral sparsification problems, from stronger notions including unit-circle approximation and singular-value approximation to weaker notions including graphical spectral sketching and effective resistance sparsification. In all of these applications, our framework produces improved results with a simpler and deterministic analysis., Comment: 32 pages

Published
2024

2. Introducing HOT3D: An Egocentric Dataset for 3D Hand and Object Tracking

Author

Banerjee, Prithviraj, Shkodrani, Sindi, Moulon, Pierre, Hampali, Shreyas, Zhang, Fan, Fountain, Jade, Miller, Edward, Basol, Selen, Newcombe, Richard, Wang, Robert, Engel, Jakob Julian, and Hodan, Tomas

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

We introduce HOT3D, a publicly available dataset for egocentric hand and object tracking in 3D. The dataset offers over 833 minutes (more than 3.7M images) of multi-view RGB/monochrome image streams showing 19 subjects interacting with 33 diverse rigid objects, multi-modal signals such as eye gaze or scene point clouds, as well as comprehensive ground truth annotations including 3D poses of objects, hands, and cameras, and 3D models of hands and objects. In addition to simple pick-up/observe/put-down actions, HOT3D contains scenarios resembling typical actions in a kitchen, office, and living room environment. The dataset is recorded by two head-mounted devices from Meta: Project Aria, a research prototype of light-weight AR/AI glasses, and Quest 3, a production VR headset sold in millions of units. Ground-truth poses were obtained by a professional motion-capture system using small optical markers attached to hands and objects. Hand annotations are provided in the UmeTrack and MANO formats and objects are represented by 3D meshes with PBR materials obtained by an in-house scanner. We aim to accelerate research on egocentric hand-object interaction by making the HOT3D dataset publicly available and by co-organizing public challenges on the dataset at ECCV 2024. The dataset can be downloaded from the project website: https://facebookresearch.github.io/hot3d/.

Published
2024

3. Analysis of Corrected Graph Convolutions

Author

Wang, Robert, Baranwal, Aseem, and Fountoulakis, Kimon

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Discrete Mathematics, Mathematics - Statistics Theory, Statistics - Machine Learning
Abstract

Machine learning for node classification on graphs is a prominent area driven by applications such as recommendation systems. State-of-the-art models often use multiple graph convolutions on the data, as empirical evidence suggests they can enhance performance. However, it has been shown empirically and theoretically, that too many graph convolutions can degrade performance significantly, a phenomenon known as oversmoothing. In this paper, we provide a rigorous theoretical analysis, based on the contextual stochastic block model (CSBM), of the performance of vanilla graph convolution from which we remove the principal eigenvector to avoid oversmoothing. We perform a spectral analysis for $k$ rounds of corrected graph convolutions, and we provide results for partial and exact classification. For partial classification, we show that each round of convolution can reduce the misclassification error exponentially up to a saturation level, after which performance does not worsen. For exact classification, we show that the separability threshold can be improved exponentially up to $O({\log{n}}/{\log\log{n}})$ corrected convolutions.

Published
2024

4. Forthcoming Spaceborne SAR Development

Author

DENG Yunkai, YU Weidong, ZHANG Heng, WANG Wei, LIU Dacheng, and WANG Robert

Subjects
spaceborne synthetic aperture radar(sar), high-resolution wide-swath, multistatic sar, multidimensional imaging, miniaturization, Electricity and magnetism, QC501-766
Abstract

Spaceborne Synthetic Aperture Radar (SAR), which can be mounted on space vehicles to collect information of the entire planet with all-day and all-weather imaging capacity, has been an indispensable device for earth observation. Currently, the technology of our spaceborne SAR has achieved a considerable technological improvement, including the resolution change from meter to submeter, the imaging mode from stripmap to azimuth beam steering like the sliding spotlight, the practical application of the multichannel approach and the conversion of single polarization into full polarization. With the development of SAR techniques, forthcoming SAR will make breakthroughs in SAR architectures, concepts, technologies and modes, for example, high-resolution wide-swath imaging, multistatic SAR, payload miniaturization and intelligence. All of these will extend the observation dimensions and obtain multidimensional data. This study focuses on the forthcoming development of spaceborne SAR.

Published
2020
Full Text
View/download PDF

5. BACQ -- Application-oriented Benchmarks for Quantum Computing

Author

Barbaresco, Frédéric, Rioux, Laurent, Labreuche, Christophe, Nowak, Michel, Olivier, Noé, Nicolazic, Damien, Hess, Olivier, Guilmin, Anne-Lise, Wang, Robert, Sassolas, Tanguy, Louise, Stéphane, Snizhko, Kyrylo, Misguich, Grégoire, Auffèves, Alexia, Whitney, Robert, Vergnaud, Emmanuelle, and Schopfer, Félicien

Subjects
Quantum Physics
Abstract

With the support of the national program on measurements, standards, and evaluation of quantum technologies MetriQs-France, a part of the French national quantum strategy, the BACQ project is dedicated to application-oriented benchmarks for quantum computing. The consortium gathering THALES, EVIDEN, an Atos business, CEA, CNRS, TERATEC, and LNE aims at establishing performance evaluation criteria of reference, meaningful for industry users.

Published
2024

6. POA Correction Method Using High-resolution Full-polarization SAR Image

Author

Sun Xiang, Song Hongjun, Wang Robert, and Li Ning

Subjects
Synthetic Aperture Radar (SAR), Full-polarization, High-resolution image, Polarization Orientation Angle (POA), Urban target decomposition, Electricity and magnetism, QC501-766
Abstract

Polarimetric decomposition in urban areas is important for monitoring the speed of city expansion and studying its ecological environmental influence. Using fully Polarimetric Synthetic Aperture Radar (PolSAR) is a method for consistent observation of large-range urban changes. In the last two decades, most research on decomposition methods have stated that Polarization Orientation Angle (POA) would affect the results of decomposition by overestimating the volume scattering contribution of urban areas. The available deorientation methods cannot rotate built-up areas with large POAs. This paper proposes an algorithm for decomposition of high-resolution urban area images based on a POA correction method. First, for high-resolution images of built-up areas, the POA changes radically pixel by pixel. An approximate assessment of urban areas can be accomplished using POA randomness. Then, to search for the true POA of large dominant POA areas (most built-up regions), the linear approximation method is used to locate POAs that can minimize cross-polarized terms. Thereby, the inaccurate decomposition that occurs by the deviation of POA can be fixed, and the accuracy of results improves. The fully PolSAR data of the Dujiangyan area in Sichuan Province, China are used to confirm the algorithm’s effectiveness. The data are acquired by an X-band airborne SAR sensor designed by the Institute of Electronics, China Academy of Sciences (IECAS).

Published
2018
Full Text
View/download PDF

7. Analysis of Time and Beam Synchronization Errors for Distributed Spaceborne SAR System

Author

Li Hangjian, Wang Robert, Deng Yunkai, Wang Wei, and Zhang Heng

Subjects
Distributed Synthetic Aperture Radar (SAR), Time synchronization, Beam synchronization, Error analysis, Electricity and magnetism, QC501-766
Abstract

Synchronization is a key problem in distributed Synthetic Aperture Radar (SAR) systems. In this paper, we perform a complex mathematical deduction and then analyze the influences of time synchronization on the SAR imaging and interferometric process. We discuss the relationship between time and phase synchronization, considering that different oscillators in separated transmitters and receivers lead to both time and phase synchronization errors. With respect to beam synchronization, we present the effects of the accuracies of beam pointing and satellite attitude on the antenna gain, based on the attitude-steering strategy, which involves azimuth weighting of the Doppler spectra for independent zero-Doppler beam steering. We also analyze the influences of beam synchronization on Doppler decorrelation, Signal-to-Noise Ratio (SNR), and overlapping swath error. We conduct simulations to validate the analysis results. Our findings provide guidance for system design.

Published
2018
Full Text
View/download PDF

8. A Novel Approach to Change Detection in SAR Images with CNN Classification

Author

Xu Zhen, Wang Robert, Li Ning, Zhang Heng, and Zhang Lei

Subjects
Synthetic Aperture Radar (SAR) image, Change detection, Convolutional Neural Networks (CNN), Electricity and magnetism, QC501-766
Abstract

This paper presents a novel Synthetic Aperture Radar (SAR)-image-change-detection method, which integrates effective-image preprocessing and Convolutional Neural Network (CNN) classification. To validate the efficiency of the proposed method, two SAR images of the same devastated region obtained by TerraSAR-X before and after the 2011 Tohoku earthquake are investigated. During image preprocessing, the image backgrounds such as mountains and water bodies are extracted and removed using Digital Elevation Model (DEM) model and Otsu’s thresholding method. A CNN is employed to automatically extract hierarchical feature representation from the data. The SAR image is then classified with the theoretically obtained features. The classification accuracies of the training and testing datasets are 98.25% and 97.86%, respectively. The changed areas between two SAR images are detected using image difference method. The accuracy and efficiency of the proposed method are validated. In addition, with other traditional methods as comparison, this paper presents change-detection results using the proposed method. Results show that the proposed method has higher accuracy in comparison with traditional change-detection methods.

Published
2017
Full Text
View/download PDF

9. CAGI, the Critical Assessment of Genome Interpretation, establishes progress and prospects for computational genetic variant interpretation methods

Author

Jain, Shantanu, Bakolitsa, Constantina, Brenner, Steven E, Radivojac, Predrag, Moult, John, Repo, Susanna, Hoskins, Roger A, Andreoletti, Gaia, Barsky, Daniel, Chellapan, Ajithavalli, Chu, Hoyin, Dabbiru, Navya, Kollipara, Naveen K, Ly, Melissa, Neumann, Andrew J, Pal, Lipika R, Odell, Eric, Pandey, Gaurav, Peters-Petrulewicz, Robin C, Srinivasan, Rajgopal, Yee, Stephen F, Yeleswarapu, Sri Jyothsna, Zuhl, Maya, Adebali, Ogun, Patra, Ayoti, Beer, Michael A, Hosur, Raghavendra, Peng, Jian, Bernard, Brady M, Berry, Michael, Dong, Shengcheng, Boyle, Alan P, Adhikari, Aashish, Chen, Jingqi, Hu, Zhiqiang, Wang, Robert, Wang, Yaqiong, Miller, Maximilian, Wang, Yanran, Bromberg, Yana, Turina, Paola, Capriotti, Emidio, Han, James J, Ozturk, Kivilcim, Carter, Hannah, Babbi, Giulia, Bovo, Samuele, Di Lena, Pietro, Martelli, Pier Luigi, Savojardo, Castrense, Casadio, Rita, Cline, Melissa S, De Baets, Greet, Bonache, Sandra, Diez, Orland, Gutierrez-Enriquez, Sara, Fernandez, Alejandro, Montalban, Gemma, Ootes, Lars, Ozkan, Selen, Padilla, Natalia, Riera, Casandra, De la Cruz, Xavier, Diekhans, Mark, Huwe, Peter J, Wei, Qiong, Xu, Qifang, Dunbrack, Roland L, Gotea, Valer, Elnitski, Laura, Margolin, Gennady, Fariselli, Piero, Kulakovskiy, Ivan V, Makeev, Vsevolod J, Penzar, Dmitry D, Vorontsov, Ilya E, Favorov, Alexander V, Forman, Julia R, Hasenahuer, Marcia, Fornasari, Maria S, Parisi, Gustavo, Avsec, Ziga, Celik, Muhammed H, Thi, Yen Duong Nguyen, Gagneur, Julien, Shi, Fang-Yuan, Edwards, Matthew D, Guo, Yuchun, Tian, Kevin, Zeng, Haoyang, Gifford, David K, Goke, Jonathan, Zaucha, Jan, Gough, Julian, Ritchie, Graham RS, Frankish, Adam, Mudge, Jonathan M, Harrow, Jennifer, Young, Erin L, and Yu, Yao

Subjects
Biological Sciences, Genetics, Human Genome, 2.1 Biological and endogenous factors, Good Health and Well Being, Humans, Computational Biology, Mutation, Missense, Phenotype, Critical Assessment of Genome Interpretation Consortium, Environmental Sciences, Information and Computing Sciences, Bioinformatics
Abstract

BackgroundThe Critical Assessment of Genome Interpretation (CAGI) aims to advance the state-of-the-art for computational prediction of genetic variant impact, particularly where relevant to disease. The five complete editions of the CAGI community experiment comprised 50 challenges, in which participants made blind predictions of phenotypes from genetic data, and these were evaluated by independent assessors.ResultsPerformance was particularly strong for clinical pathogenic variants, including some difficult-to-diagnose cases, and extends to interpretation of cancer-related variants. Missense variant interpretation methods were able to estimate biochemical effects with increasing accuracy. Assessment of methods for regulatory variants and complex trait disease risk was less definitive and indicates performance potentially suitable for auxiliary use in the clinic.ConclusionsResults show that while current methods are imperfect, they have major utility for research and clinical applications. Emerging methods and increasingly large, robust datasets for training and assessment promise further progress ahead.

Published
2024

10. Range Ambiguity Suppression Approach for Quad-pol SAR Systems Based on Modified Azimuth Phase Coding

Author

Zhu Xiaojing, Li Fei, Wang Robert, Wang Wei, and Sun Xiang

Subjects
Synthetic Aperture Radar (SAR), Quadrature-polarimetric (quad-pol), Hybrid-polarimetric (hybrid-pol), Range ambiguity, Modified Azimuth Phase Coding (MAPC), Wiener filter, Electricity and magnetism, QC501-766
Abstract

For conventional quadrature-polarimetric (quad-pol) Synthetic Aperture Radar (SAR) systems, as cross-polarized (cross-pol) channels are influenced by the strong co-polarized (co-pol) range ambiguous returns, the range ambiguity levels of cross-polchannels are markedly reduced, which severely restricts the unambiguous swaths. A novel transmission scheme called a hybrid-polarimetric (hybrid-pol) mode is introduced to enhance the range ambiguity levels of cross-pol channels. This scheme improves the performance of cross-pol channels with regards to range ambiguity but deteriorates that of co-pol channels. Therefore, to further enhance the range ambiguity levels of quad-pol SAR systems, the Modified Azimuth Phase Coding (MAPC) technique based on hybrid-pol SAR systems is proposed in this study. By taking advantage of the MAPC modulation/ demodulation, the power of range ambiguities is transferred to the azimuth that is filtered by an optimized Wiener filter in the Doppler domain. The simulation results validate that the MAPC technique can markedly eliminate the range ambiguity of quad-pol SAR systems and extend the unambiguous swaths.

Published
2017
Full Text
View/download PDF

11. High-resolution Slide Spotlight SAR Imaging by BP Algorithm and Heterogeneous Parallel Implementation

Author

Tang Jiangwen, Deng Yunkai, Wang Robert, Zhao Shuo, and Li Ning

Subjects
Synthetic Aperture Radar (SAR), Slide spotlight, Back-Projection (BP) algorithm, Heterogeneous parallel computing, Electricity and magnetism, QC501-766
Abstract

High-resolution synthetic aperture radar presents a significant challenge to imaging algorithms and computing power. Slide spotlight is an important mode that has both high resolution and wide azimuth swath. Generally, in the slide spotlight mode, the performance of conventional frequency domain imaging algorithms degrades because of orbit curvature, the time-variant azimuth chirp rate, and other factors. We adopt the Back-Projection (BP) algorithm in this study to counteract this limitation. We also propose a CPU/GPU heterogeneous BP algorithm to deal with the high computing complexity O(N3) of the BP algorithm. This heterogeneous BP algorithm makes full use of computing resources and accelerates imaging progress, and the design of a scheduling thread improves the flexibility of the algorithm.

Published
2017
Full Text
View/download PDF

12. Signal Reconstruction Approach for Multichannel SAR in Azimuth Based on Multiframe Super resolution

Author

Zhao Qingchao, Zhang Yi, Wang Robert, Wang Wei, and Wang Xiangyu

Subjects
Synthetic Aperture Radar (SAR), Azimuth multichannel, High resolution wide swath, Pulse Repetition Frequency (PRF), Nonuniform sampling, Multiframe super-resolution, Electricity and magnetism, QC501-766
Abstract

To achieve high-resolution wide-swath imaging, the use of multichannel techniques in azimuth is effective for spaceborne Synthetic Aperture Radar (SAR). For azimuth multichannel systems, the signal in azimuth is nonuniformly sampled if the uniform sampling condition related to Pulse Repetition Frequency (PRF) is not satisfied, which makes it important to reconstruct the azimuth signal prior to image formation. In this study, to solve the azimuth signal reconstruction problem in multichannel SAR, we propose the innovative use of a multiframe super-resolution method in Digital Image Processing (DIP) and summarize the general multiframe super-resolution process. Our simulation results and real data experiments verify the effectiveness of the proposed method, which demonstrates some advantages in complexity performance. By establishing linkages between the problem of signal reconstruction of nonuniformly sampled signals and the multiframe superresolution concept, we provide a new approach to this traditional signal reconstruction problem.

Published
2017
Full Text
View/download PDF

13. Fast Algorithms for Directed Graph Partitioning Using Flows and Reweighted Eigenvalues

Author

Lau, Lap Chi, Tung, Kam Chuen, and Wang, Robert

Subjects
Computer Science - Data Structures and Algorithms, Computer Science - Discrete Mathematics, Computer Science - Machine Learning, Mathematics - Combinatorics
Abstract

We consider a new semidefinite programming relaxation for directed edge expansion, which is obtained by adding triangle inequalities to the reweighted eigenvalue formulation. Applying the matrix multiplicative weight update method to this relaxation, we derive almost linear-time algorithms to achieve $O(\sqrt{\log{n}})$-approximation and Cheeger-type guarantee for directed edge expansion, as well as an improved cut-matching game for directed graphs. This provides a primal-dual flow-based framework to obtain the best known algorithms for directed graph partitioning. The same approach also works for vertex expansion and for hypergraphs, providing a simple and unified approach to achieve the best known results for different expansion problems and different algorithmic techniques.

Published
2023

14. Experimental Design for Any $p$-Norm

Author

Lau, Lap Chi, Wang, Robert, and Zhou, Hong

Subjects
Computer Science - Data Structures and Algorithms, Computer Science - Machine Learning, Statistics - Computation, Statistics - Machine Learning
Abstract

We consider a general $p$-norm objective for experimental design problems that captures some well-studied objectives (D/A/E-design) as special cases. We prove that a randomized local search approach provides a unified algorithm to solve this problem for all $p$. This provides the first approximation algorithm for the general $p$-norm objective, and a nice interpolation of the best known bounds of the special cases., Comment: 29 pages

Published
2023

16. IRIS: Discovery of cancer immunotherapy targets arising from pre-mRNA alternative splicing

Author

Pan, Yang, Phillips, John W, Zhang, Beatrice D, Noguchi, Miyako, Kutschera, Eric, McLaughlin, Jami, Nesterenko, Pavlo A, Mao, Zhiyuan, Bangayan, Nathanael J, Wang, Robert, Tran, Wendy, Yang, Harry T, Wang, Yuanyuan, Xu, Yang, Obusan, Matthew B, Cheng, Donghui, Lee, Alex H, Kadash-Edmondson, Kathryn E, Champhekar, Ameya, Puig-Saus, Cristina, Ribas, Antoni, Prins, Robert M, Seet, Christopher S, Crooks, Gay M, Witte, Owen N, and Xing, Yi

Subjects
Biomedical and Clinical Sciences, Oncology and Carcinogenesis, Immunology, Cancer Genomics, Biotechnology, Immunotherapy, Immunization, Cancer, Vaccine Related, Human Genome, Genetics, 5.1 Pharmaceuticals, Development of treatments and therapeutic interventions, Good Health and Well Being, Male, Humans, RNA Precursors, Alternative Splicing, Leukocytes, Mononuclear, Receptors, Antigen, T-Cell, Epitopes, T-Lymphocyte, Antigens, Neoplasm, Peptides, Neoplasms, RNA splicing, immunotherapy, T cell receptors
Abstract

Alternative splicing (AS) is prevalent in cancer, generating an extensive but largely unexplored repertoire of novel immunotherapy targets. We describe Isoform peptides from RNA splicing for Immunotherapy target Screening (IRIS), a computational platform capable of discovering AS-derived tumor antigens (TAs) for T cell receptor (TCR) and chimeric antigen receptor T cell (CAR-T) therapies. IRIS leverages large-scale tumor and normal transcriptome data and incorporates multiple screening approaches to discover AS-derived TAs with tumor-associated or tumor-specific expression. In a proof-of-concept analysis integrating transcriptomics and immunopeptidomics data, we showed that hundreds of IRIS-predicted TCR targets are presented by human leukocyte antigen (HLA) molecules. We applied IRIS to RNA-seq data of neuroendocrine prostate cancer (NEPC). From 2,939 NEPC-associated AS events, IRIS predicted 1,651 epitopes from 808 events as potential TCR targets for two common HLA types (A*02:01 and A*03:01). A more stringent screening test prioritized 48 epitopes from 20 events with "neoantigen-like" NEPC-specific expression. Predicted epitopes are often encoded by microexons of ≤30 nucleotides. To validate the immunogenicity and T cell recognition of IRIS-predicted TCR epitopes, we performed in vitro T cell priming in combination with single-cell TCR sequencing. Seven TCRs transduced into human peripheral blood mononuclear cells (PBMCs) showed high activity against individual IRIS-predicted epitopes, providing strong evidence of isolated TCRs reactive to AS-derived peptides. One selected TCR showed efficient cytotoxicity against target cells expressing the target peptide. Our study illustrates the contribution of AS to the TA repertoire of cancer cells and demonstrates the utility of IRIS for discovering AS-derived TAs and expanding cancer immunotherapies.

Published
2023

18. Cheeger Inequalities for Directed Graphs and Hypergraphs Using Reweighted Eigenvalues

Author

Lau, Lap Chi, Tung, Kam Chuen, and Wang, Robert

Subjects
Computer Science - Data Structures and Algorithms, Computer Science - Discrete Mathematics, Computer Science - Machine Learning, Mathematics - Combinatorics, Mathematics - Probability
Abstract

We derive Cheeger inequalities for directed graphs and hypergraphs using the reweighted eigenvalue approach that was recently developed for vertex expansion in undirected graphs [OZ22,KLT22,JPV22]. The goal is to develop a new spectral theory for directed graphs and an alternative spectral theory for hypergraphs. The first main result is a Cheeger inequality relating the vertex expansion $\vec{\psi}(G)$ of a directed graph $G$ to the vertex-capacitated maximum reweighted second eigenvalue $\vec{\lambda}_2^{v*}$: \[ \vec{\lambda}_2^{v*} \lesssim \vec{\psi}(G) \lesssim \sqrt{\vec{\lambda}_2^{v*} \cdot \log (\Delta/\vec{\lambda}_2^{v*})}. \] This provides a combinatorial characterization of the fastest mixing time of a directed graph by vertex expansion, and builds a new connection between reweighted eigenvalued, vertex expansion, and fastest mixing time for directed graphs. The second main result is a stronger Cheeger inequality relating the edge conductance $\vec{\phi}(G)$ of a directed graph $G$ to the edge-capacitated maximum reweighted second eigenvalue $\vec{\lambda}_2^{e*}$: \[ \vec{\lambda}_2^{e*} \lesssim \vec{\phi}(G) \lesssim \sqrt{\vec{\lambda}_2^{e*} \cdot \log (1/\vec{\lambda}_2^{e*})}. \] This provides a certificate for a directed graph to be an expander and a spectral algorithm to find a sparse cut in a directed graph, playing a similar role as Cheeger's inequality in certifying graph expansion and in the spectral partitioning algorithm for undirected graphs. We also use this reweighted eigenvalue approach to derive the improved Cheeger inequality for directed graphs, and furthermore to derive several Cheeger inequalities for hypergraphs that match and improve the existing results in [Lou15,CLTZ18]. These are supporting results that this provides a unifying approach to lift the spectral theory for undirected graphs to more general settings., Comment: 51 pages, 3 figures

Published
2022

19. UmeTrack: Unified multi-view end-to-end hand tracking for VR

Author

Han, Shangchen, Wu, Po-chen, Zhang, Yubo, Liu, Beibei, Zhang, Linguang, Wang, Zheng, Si, Weiguang, Zhang, Peizhao, Cai, Yujun, Hodan, Tomas, Cabezas, Randi, Tran, Luan, Akbay, Muzaffer, Yu, Tsz-Ho, Keskin, Cem, and Wang, Robert

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Real-time tracking of 3D hand pose in world space is a challenging problem and plays an important role in VR interaction. Existing work in this space are limited to either producing root-relative (versus world space) 3D pose or rely on multiple stages such as generating heatmaps and kinematic optimization to obtain 3D pose. Moreover, the typical VR scenario, which involves multi-view tracking from wide \ac{fov} cameras is seldom addressed by these methods. In this paper, we present a unified end-to-end differentiable framework for multi-view, multi-frame hand tracking that directly predicts 3D hand pose in world space. We demonstrate the benefits of end-to-end differentiabilty by extending our framework with downstream tasks such as jitter reduction and pinch prediction. To demonstrate the efficacy of our model, we further present a new large-scale egocentric hand pose dataset that consists of both real and synthetic data. Experiments show that our system trained on this dataset handles various challenging interactive motions, and has been successfully applied to real-time VR applications., Comment: SIGGRAPH Asia 2022 Conference Papers, 8 pages

Published
2022
Full Text
View/download PDF

20. MotionDeltaCNN: Sparse CNN Inference of Frame Differences in Moving Camera Videos

Author

Parger, Mathias, Tang, Chengcheng, Neff, Thomas, Twigg, Christopher D., Keskin, Cem, Wang, Robert, and Steinberger, Markus

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning
Abstract

Convolutional neural network inference on video input is computationally expensive and requires high memory bandwidth. Recently, DeltaCNN managed to reduce the cost by only processing pixels with significant updates over the previous frame. However, DeltaCNN relies on static camera input. Moving cameras add new challenges in how to fuse newly unveiled image regions with already processed regions efficiently to minimize the update rate - without increasing memory overhead and without knowing the camera extrinsics of future frames. In this work, we propose MotionDeltaCNN, a sparse CNN inference framework that supports moving cameras. We introduce spherical buffers and padded convolutions to enable seamless fusion of newly unveiled regions and previously processed regions -- without increasing memory footprint. Our evaluation shows that we outperform DeltaCNN by up to 90% for moving camera videos.

Published
2022

21. Neural Correspondence Field for Object Pose Estimation

Author

Huang, Lin, Hodan, Tomas, Ma, Lingni, Zhang, Linguang, Tran, Luan, Twigg, Christopher, Wu, Po-Chen, Yuan, Junsong, Keskin, Cem, and Wang, Robert

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, Computer Science - Machine Learning, Computer Science - Robotics
Abstract

We propose a method for estimating the 6DoF pose of a rigid object with an available 3D model from a single RGB image. Unlike classical correspondence-based methods which predict 3D object coordinates at pixels of the input image, the proposed method predicts 3D object coordinates at 3D query points sampled in the camera frustum. The move from pixels to 3D points, which is inspired by recent PIFu-style methods for 3D reconstruction, enables reasoning about the whole object, including its (self-)occluded parts. For a 3D query point associated with a pixel-aligned image feature, we train a fully-connected neural network to predict: (i) the corresponding 3D object coordinates, and (ii) the signed distance to the object surface, with the first defined only for query points in the surface vicinity. We call the mapping realized by this network as Neural Correspondence Field. The object pose is then robustly estimated from the predicted 3D-3D correspondences by the Kabsch-RANSAC algorithm. The proposed method achieves state-of-the-art results on three BOP datasets and is shown superior especially in challenging cases with occlusion. The project website is at: linhuang17.github.io/NCF., Comment: Accepted to ECCV 2022

Published
2022

22. Assembly101: A Large-Scale Multi-View Video Dataset for Understanding Procedural Activities

Author

Sener, Fadime, Chatterjee, Dibyadip, Shelepov, Daniel, He, Kun, Singhania, Dipika, Wang, Robert, and Yao, Angela

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Assembly101 is a new procedural activity dataset featuring 4321 videos of people assembling and disassembling 101 "take-apart" toy vehicles. Participants work without fixed instructions, and the sequences feature rich and natural variations in action ordering, mistakes, and corrections. Assembly101 is the first multi-view action dataset, with simultaneous static (8) and egocentric (4) recordings. Sequences are annotated with more than 100K coarse and 1M fine-grained action segments, and 18M 3D hand poses. We benchmark on three action understanding tasks: recognition, anticipation and temporal segmentation. Additionally, we propose a novel task of detecting mistakes. The unique recording format and rich set of annotations allow us to investigate generalization to new toys, cross-view transfer, long-tailed distributions, and pose vs. appearance. We envision that Assembly101 will serve as a new challenge to investigate various activity understanding problems., Comment: CVPR 2022, https://assembly-101.github.io/

Published
2022

23. DeltaCNN: End-to-End CNN Inference of Sparse Frame Differences in Videos

Author

Parger, Mathias, Tang, Chengcheng, Twigg, Christopher D., Keskin, Cem, Wang, Robert, and Steinberger, Markus

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning
Abstract

Convolutional neural network inference on video data requires powerful hardware for real-time processing. Given the inherent coherence across consecutive frames, large parts of a video typically change little. By skipping identical image regions and truncating insignificant pixel updates, computational redundancy can in theory be reduced significantly. However, these theoretical savings have been difficult to translate into practice, as sparse updates hamper computational consistency and memory access coherence; which are key for efficiency on real hardware. With DeltaCNN, we present a sparse convolutional neural network framework that enables sparse frame-by-frame updates to accelerate video inference in practice. We provide sparse implementations for all typical CNN layers and propagate sparse feature updates end-to-end - without accumulating errors over time. DeltaCNN is applicable to all convolutional neural networks without retraining. To the best of our knowledge, we are the first to significantly outperform the dense reference, cuDNN, in practical settings, achieving speedups of up to 7x with only marginal differences in accuracy., Comment: CVPR 2022

Published
2022

24. Assessment of Visual Function with Cotoretigene Toliparvovec in X-Linked Retinitis Pigmentosa in the Randomized XIRIUS Phase 2/3 Study

Author

MacLaren, Robert, Aleman, Tomas, Birch, David, Jalil, Assad, Lotery, Andrew, Lam, Byron, Pennesi, Mark, Kay, Christine N., Yusuf, Imram H., Cehajic Kapetanovic, Jasmina, Jolly, Jasleen K., Josan, Amandeep S., Taylor, Laura J., Xue, Kanmin, Nanda, Anika, Buckley, Thomas, Salvetti, Anna Paola, Thulasidharan, Suresh, Kurc, Miguel, Khandhadia, Samir, Dias, Karla Orsine Murta, Baltmr, Abeir, Mandal, Nakul, Tsokolas, Georgios, Ivanova, Tsveta, EL-Faouri, Muhannd, Shaarawy, Sherif, Black, Graeme, Davis, Janet L., Gregori, Ninel Z., Mendoza-Santiesteban, Carlos E., Lauer, Andreas K., Yang, Paul, Bailey, Steven, Spencer, Rand, Fish, Gary E., Wang, Robert, Chong, Deborah, Abbey, Ashkan, Anand, Rajiv, MaGuire, Albert A., Roseman, Robert L., Hazariwala, Kaushik M., Parrott, Brandon, Lam, Byron L., Pennesi, Mark E., Panda, Sushil, Gow, James A., Zhao, Guolin, and MacLaren, Robert E.

Published
2024
Full Text
View/download PDF

29. The impact of perioperative stroke and delirium on outcomes after surgical aortic valve replacement

Author

Miller, Marissa A., Taddei-Peters, Wendy C., Jeffries, Neal O., Buxton, Dennis, Geller, Nancy L., Gordon, David, Burke, Catherine, Lee, Albert, Smith, Tyrone, Moy, Claudia S., Gombos, Ilana Kogan, Weisel, Richard, Gardner, Timothy J., O'Gara, Patrick T., Rose, Eric A., Gelijns, Annetine C., Parides, Michael K., Ascheim, Deborah D., Moskowitz, Alan J., Bagiella, Emilia, Moquete, Ellen, Shah, Kinjal, Overbey, Jessica R., Pan, Stephanie, Chang, Helena, Chase, Melissa, Goldfarb, Seth, Gupta, Lopa, Kirkwood, Katherine, Dobrev, Edlira, Levitan, Ron, O'Sullivan, Karen, Santos, Milerva, Ye, Xia, Mack, Michael, Winkle, Rachelle, Boswell, Haley, Fenlon, Amanda, Johnson, Melissa, Jones, Jessica, Kolb, Megan, Lam, Sarah, Miranda, Lucy, Ward, Jackie, Whitman, Renessa, Zingler, Brittany, Ryan, William, Smith, Robert L., Grayburn, Paul, Nosnik, Pedro, Gillinov, A. Marc, Blackstone, Eugene H., Moazami, Nader, Starling, Randall C., Barzilai, Benico, Grimm, Richard A., Soltesz, Edward G., Katzan, Irene, Strippy, Brian, Smith, Shoi, Garcia, Michelle, Alice bowman, Mary, Geither, Carrie, Wang, Robert, Argenziano, Michael, Borger, Michael, Takayama, Hiroo, Leon, Martin B., Goldsmith, Lyn, Schwartz, Allan, Sookraj, Nadia, McCright-Gill, Talaya, Sreekanth, Sowmya, McCullough, Jock N., Iribarne, Alexander, DeSimone, Joseph P., DiScipio, Anthony W., Stokes, Henry, Ivany, Amanda St., Petty, Gaylin, Smith, Peter K., Alexander, John H., Milano, Carmelo A., Glower, Donald D., Huber, Joel, Morganlander, Joel, Mathew, Joseph P., Welsh, Stacey, Casalinova, Sarah, Johnson, Victoria, Lane, Kathleen, Smith, Derek, Tipton, Greg, Berry, Mark F., Williams, Judson B., Englum, Brian, Hartwig, Matthew, Thourani, Vinod H., Guyton, Robert, Lattouf, Omar, Chen, Edward, Vega, J. David, Baer, Jefferson, Nguyen, Duc, Halkos, Michael, Baio, Kim, Prince, Tamara, Cook, Natascha, Neill, Alexis A., Voisine, Pierre, Senechal, Mario, Dagenais, François, Laforce, Robert, Jr., O'Connor, Kim, Dussault, Gladys, Caouette, Manon, Tremblay, Hugo, Gagne, Nathalie, Dumont, Julie, Landry, Patricia, Groh, Mark A., Trichon, Benjamin H., Binns, Oliver A., Ely, Stephen W., Johnson, Alan M., Hansen, Todd H., Short, John G., Taylor, Reid D., Mangusan, Ralph, Nanney, Tracy, Aubart, Holly, Cross, Kristin, McPeters, Leslie, Riggsbee, Christina, Rixey, Lucy, Michler, Robert E., DeRose, Joseph J., Jr., Goldstein, Daniel J., Bello, Ricardo A., Taub, Cynthia, Spevack, Daniel, Kirchoff, Kathryn, Meli, Rebecca, Garcia, Juan, Goldenberg, Jon, Kealy, Lauren, Perrault, Louis P., Bouchard, Denis, Tanguay, Jean François, O'Meara, Eileen, Lacharité, Jonathan, Robichaud, Sophie, Horvath, Keith A., Corcoran, Philip C., Siegenthaler, Michael P., Murphy, Mandy, Iraola, Margaret, Greenberg, Ann, Kumkumian, Greg, Milner, Mark, Nadareishvili, Zurab, Whitson, Bryan A., Hasan, Ayesha, McDavid, Asia, Fadorsen, Denise, Ouzounian, Maral, Yau, Terry, Farkouh, Michael, Woo, Anna, Cusimano, Robert James, David, Tirone, Feindel, Christopher, Fumakia, Nishit, Christie, Shakira, Mullen, John C., Bissonauth, Asvina, Hripko, Alexandra, Gammie, James S., Noor, Zahid, Mackowick, Kristen, Deasey, Stephanie, Al-Suqi, Manal, Collins, Julia, Acker, Michael A., Messé, Steven, Kirkpatrick, James, Mayer, Mary Lou, McDonald, Caitlin, Fok, Holley, Maffei, Breanna, Cresse, Stephen, Gepty, Christine, Bowdish, Michael, Starnes, Vaughn A., Shavalle, David, Heck, Christi, Hackmann, Amy, Baker, Craig, Fleischman, Fernando, Cunningham, Mark, Lozano, Edward, Hernandez, Michelle, Ailawadi, Gorav, Kron, Irving L., Johnston, Karen, Ghanta, Ravi K., Dent, John M., Kern, John, Yarboro, Leora, Ragosta, Michael, Annex, Brian, Bergin, Jim, Burks, Sandra, Cosner, Mike, Green, China, Loya, Samantha, Kim, Hye Ryun, Bull, David A., Desvigne-Nickens, Patrice, Dixon, Dennis O., Gottesman, Rebecca, Haigney, Mark, Holubkov, Richard, Iadecola, Constantino, Jacobs, Alice, Meslin, Eric M., Murkin, John M., Spertus, John A., Sellke, Frank, McDonald, Cheryl L., Canty, John, Dickert, Neal, Ikonomidis, John S., Kim, KyungMann, Williams, David O., Yancy, Clyde W., Chaturvedi, Seemant, Chimowitz, Marc, Fang, James C., Richenbacher, Wayne, Rao, Vivek, Furie, Karen L., Miller, Rachel, Cook, Jennifer, D'Alessandro, David, Han, Frederick, Pinney, Sean, Walsh, Mary N., Greer, David, Ishida, Koto, Stapf, Christian, Hung, Judy, Zeng, Xin, Hung, David, Satitthummanid, Sudarat, Billelo, Michel, Davatzikos, Christos, Erus, Guray, Karpf, Lauren, Desiderio, Lisa, Browndyke, Jeffrey N., James, Michael L., Toulgoat-Dubois, Yanne, Brassard, Rachele, Virmanu, Renu, Romero, Maria E., Braumann, Ryan, Messé, Steven R., Mack, Michael J., Southerland, Andrew M., Moy, Claudia Scala, and Bowdish, Michael E.

Published
2024
Full Text
View/download PDF

30. UNOC: Understanding Occlusion for Embodied Presence in Virtual Reality

Author

Parger, Mathias, Tang, Chengcheng, Xu, Yuanlu, Twigg, Christopher, Tao, Lingling, Li, Yijing, Wang, Robert, and Steinberger, Markus

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning
Abstract

Tracking body and hand motions in the 3D space is essential for social and self-presence in augmented and virtual environments. Unlike the popular 3D pose estimation setting, the problem is often formulated as inside-out tracking based on embodied perception (e.g., egocentric cameras, handheld sensors). In this paper, we propose a new data-driven framework for inside-out body tracking, targeting challenges of omnipresent occlusions in optimization-based methods (e.g., inverse kinematics solvers). We first collect a large-scale motion capture dataset with both body and finger motions using optical markers and inertial sensors. This dataset focuses on social scenarios and captures ground truth poses under self-occlusions and body-hand interactions. We then simulate the occlusion patterns in head-mounted camera views on the captured ground truth using a ray casting algorithm and learn a deep neural network to infer the occluded body parts. In the experiments, we show that our method is able to generate high-fidelity embodied poses by applying the proposed method on the task of real-time inside-out body tracking, finger motion synthesis, and 3-point inverse kinematics.

Published
2020

31. Lightweight Multi-View 3D Pose Estimation through Camera-Disentangled Representation

Author

Remelli, Edoardo, Han, Shangchen, Honari, Sina, Fua, Pascal, and Wang, Robert

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

We present a lightweight solution to recover 3D pose from multi-view images captured with spatially calibrated cameras. Building upon recent advances in interpretable representation learning, we exploit 3D geometry to fuse input images into a unified latent representation of pose, which is disentangled from camera view-points. This allows us to reason effectively about 3D pose across different views without using compute-intensive volumetric grids. Our architecture then conditions the learned representation on camera projection operators to produce accurate per-view 2d detections, that can be simply lifted to 3D via a differentiable Direct Linear Transform (DLT) layer. In order to do it efficiently, we propose a novel implementation of DLT that is orders of magnitude faster on GPU architectures than standard SVD-based triangulation methods. We evaluate our approach on two large-scale human pose datasets (H36M and Total Capture): our method outperforms or performs comparably to the state-of-the-art volumetric methods, while, unlike them, yielding real-time performance.

Published
2020

32. RNA Dysregulation: An Expanding Source of Cancer Immunotherapy Targets

Author

Pan, Yang, Kadash-Edmondson, Kathryn E, Wang, Robert, Phillips, John, Liu, Song, Ribas, Antoni, Aplenc, Richard, Witte, Owen N, and Xing, Yi

Subjects
Biomedical and Clinical Sciences, Oncology and Carcinogenesis, Immunology, Rare Diseases, Genetics, Immunization, Biotechnology, Cancer, Vaccine Related, Humans, Immunotherapy, Neoplasms, Proteome, RNA, Transcriptome, CAR-T, RNA processing, RNA-seq, TCR, cancer immunotherapy, proteogenomics, proteome, transcriptome, tumor antigen, Biological Sciences, Medical and Health Sciences, Pharmacology & Pharmacy, Pharmacology and pharmaceutical sciences
Abstract

Cancer transcriptomes frequently exhibit RNA dysregulation. As the resulting aberrant transcripts may be translated into cancer-specific proteins, there is growing interest in exploiting RNA dysregulation as a source of tumor antigens (TAs) and thus novel immunotherapy targets. Recent advances in high-throughput technologies and rapid accumulation of multiomic cancer profiling data in public repositories have provided opportunities to systematically characterize RNA dysregulation in cancer and identify antigen targets for immunotherapy. However, given the complexity of cancer transcriptomes and proteomes, important conceptual and technological challenges exist. Here, we highlight the expanding repertoire of TAs arising from RNA dysregulation and introduce multiomic and big data strategies for identifying optimal immunotherapy targets. We discuss extant barriers for translating these targets into effective therapies as well as the implications for future research.

Published
2021

35. DnaJ homolog Hdj2 Facilitates Japanese Encephalitis Virus Replication

Author

Hung Chun-Yu, Chong Ka-Man, Huang Yu-Ru, Wang Robert, Ke Zhi-Long, and Chang Ruey-Yi

Subjects
JEV, Hsp40, Hdj2, and RdRp, Infectious and parasitic diseases, RC109-216
Abstract

Abstract Background Japanese encephalitis virus (JEV) is a member of the mosquito-borne Flaviviridae family of viruses that causes human encephalitis. Upon infection of a new host, replication of viral RNA involves not only the viral RNA-dependent RNA polymerase (RdRp), but also host proteins. Host factors involved in JEV replication are not well characterized. Results We identified Hdj2, a heat-shock protein 40 (Hsp40)/DnaJ homolog, from a mouse brain cDNA library interacting with JEV nonstructural protein 5 (NS5) encoding viral RdRp using yeast two-hybrid system. Specific interaction of Hdj2 with NS5 was confirmed by coimmunoprecipitation and colocalization in JEV-infected cells. Overexpression of Hdj2 in JEV-infected cells led to an increase of RNA synthesis, and the virus titer was elevated approximately 4.5- to 10-fold. Knocking down of Hdj2 by siRNA reduced the virus production significantly. Conclusions We conclude that Hdj2 directly associates with JEV NS5 and facilitates viral replication. This study is the first to demonstrate Hdj2 involved in JEV replication, providing insight into a potential therapeutic target and cell-based vaccine development of JEV infection.

Published
2011
Full Text
View/download PDF

36. High-risk human papillomavirus (HPV) screening and detection in healthy patient saliva samples: a pilot study

Author

Wang Robert C, Barber Annabel E, Mauer Jarom, Martin Daniel, Falk Jesse, Catmull Jeremy, Bullen Ryan, Williams-Cocks Shelley J, Turner Deidre O, Gerstenberger Shawn L, and Kingsley Karl

Subjects
Dentistry, RK1-715
Abstract

Abstract Background The human papillomaviruses (HPV) are a large family of non-enveloped DNA viruses, mainly associated with cervical cancers. Recent epidemiologic evidence has suggested that HPV may be an independent risk factor for oropharyngeal cancers. Evidence now suggests HPV may modulate the malignancy process in some tobacco- and alcohol-induced oropharynx tumors, but might also be the primary oncogenic factor for inducing carcinogenesis among some non-smokers. More evidence, however, is needed regarding oral HPV prevalence among healthy adults to estimate risk. The goal of this study was to perform an HPV screening of normal healthy adults to assess oral HPV prevalence. Methods Healthy adult patients at a US dental school were selected to participate in this pilot study. DNA was isolated from saliva samples and screened for high-risk HPV strains HPV16 and HPV18 and further processed using qPCR for quantification and to confirm analytical sensitivity and specificity. Results Chi-square analysis revealed the patient sample was representative of the general clinic population with respect to gender, race and age (p < 0.05). Four patient samples were found to harbor HPV16 DNA, representing 2.6% of the total (n = 151). Three of the four HPV16-positive samples were from patients under 65 years of age and all four were female and Hispanic (non-White). No samples tested positive for HPV18. Conclusions The successful recruitment and screening of healthy adult patients revealed HPV16, but not HPV18, was present in a small subset. These results provide new information about oral HPV status, which may help to contextualize results from other studies that demonstrate oral cancer rates have risen in the US among both females and minorities and in some geographic areas that are not solely explained by rates of tobacco and alcohol use. The results of this study may be of significant value to further our understanding of oral health and disease risk, as well as to help design future studies exploring the role of other factors that influence oral HPV exposure, as well as the short- and long-term consequences of oral HPV infection.

Published
2011
Full Text
View/download PDF

38. Japanese encephalitis virus co-opts the ER-stress response protein GRP78 for viral infectivity

Author

Tsai Meng-Chieh, Hung Chun-Yu, Chang Chung-Ming, Wu Yi-Ping, Schuyler Scott C, and Wang Robert

Subjects
Infectious and parasitic diseases, RC109-216
Abstract

Abstract The serum-free medium from Japanese encephalitis virus (JEV) infected Baby Hamster Kidney-21 (BHK-21) cell cultures was analyzed by liquid chromatography tandem mass spectrometry (LC-MS) to identify host proteins that were secreted upon viral infection. Five proteins were identified, including the molecular chaperones Hsp90, GRP78, and Hsp70. The functional role of GRP78 in the JEV life cycle was then investigated. Co-migration of GRP78 with JEV particles in sucrose density gradients was observed and co-localization of viral E protein with GRP78 was detected by immunofluorescence analysis in vivo. Knockdown of GRP78 expression by siRNA did not effect viral RNA replication, but did impair mature viral production. Mature viruses that do not co-fractionate with GPR78 displayed a significant decrease in viral infectivity. Our results support the hypothesis that JEV co-opts host cell GPR78 for use in viral maturation and in subsequent cellular infections.

Published
2011
Full Text
View/download PDF

39. Prediction for 2-Year Vision Outcomes Using Early Morphologic and Functional Responses in the Comparison of Age-related Macular Degeneration Treatments Trials

Author

Williams, David F., Beardsley, Sara, Bennett, Steven, Cantrill, Herbert, Chan-Tram, Carmen, Cheshier, Holly, Damato, Kathyrn, Davies, John, Dev, Sundeep, Enloe, Julianne, Follano, Gennaro, Gilbert, Peggy, Johnson, Jill, Jones, Tori, Mayleben, Lisa, Mittra, Robert, Moos, Martha, Neist, Ryan, Oestreich, Neal, Quiram, Polly, Ramsay, Robert, Ryan, Edwin, Schindeldecker, Stephanie, Snater, John, Steele, Trenise, Selders, Dwight, Tonsfeldt, Jessica, Valardi, Shelly, Fish, Gary Edd, Aguado, Hank A., Arceneaux, Sally, Arnwine, Jean, Bell, Kim, Bell, Tina, Boleman, Bob, Bradley, Patricia, Callanan, David, Coors, Lori, Creighton, Jodi, Crew, Timothy, Cummings, Kimberly, Dock, Christopher, Duignan, Karen, Fuller, Dwain, Gray, Keith, Hendrix, Betsy, Hesse, Nicholas, Jaramillo, Diana, Jost, Bradley, Lash, Sandy, Lonsdale, Laura, Mackens, Michael, Mutz, Karin, Potts, Michael, Sanchez, Brenda, Snyder, William, Solley, Wayne, Tarter, Carrie, Wang, Robert, Williams, Patrick, Perkins, Stephen L., Anderson, Nicholas, Arnold, Ann, Blais, Paul, Googe, Joseph, Higdon, Tina T., Hunt, Cecile, Johnson, Mary, Miller, James, Moore, Misty, Morris, Charity K., Morris, Christopher, Oelrich, Sarah, Oliver, Kristina, Seitz, Vicky, Whetstone, Jerry, Doft, Bernard H., Bedel, Jay, Bergren, Robert, Borthwick, Ann, Conrad, Paul, Fec, Amanda, Fulwylie, Christina, Ingram, Willia, Latham, Shawnique, Lester, Gina, Liu, Judy, Lobes, Louis, Lucko, Nicole M., Mechling, Holly, Merlotti, Lori, McBroom, Keith, Olsen, Karl, Puskas, Danielle, Rath, Pamela, Schmucker, Maria, Schueckler, Lynn, Schultz, Christina, Shultz, Heather, Steinberg, David, Vyas, Avni, Whale, Kim, Yeckel, Kimberly, Orth, David H., Arredondo, Linda S., Brown, Susan, Ciscato, Barbara J., Civantos, Joseph M., Figliulo, Celeste, Hasan, Sohail, Kosinski, Belinda, Muir, Dan, Nelson, Kiersten, Packo, Kirk, Pollack, John S., Rezaei, Kourous, Shelton, Gina, Townsend-Patrick, Shannya, Walsh, Marian, McDonald, H. Richard, Ansari, Nina, Bye, Amanda, Fu, Arthur D., Grout, Sean, Indermill, Chad, Johnson, Robert N., Jumper, J. Michael, Linares, Silvia, Lujan, Brandon J., Munden, Ames, Persons, Meredith, Rodriguez, Rosa, Rose, Jennifer M., Teske, Brandi, Urias, Yesmin, Young, Stephen, Dreyer, Richard F., Daniel, Howard, Connaughton, Michele, Handelman, Irvin, Hobbs, Stephen, Hoerner, Christine, Hudson, Dawn, Kopfer, Marcia, Lee, Michael, Lemley, Craig, Logan, Joe, Ma, Colin, Mallet, Christophe, Milliron, Amanda, Peters, Mark, Wohlsein, Harry, Pearlman, Joel A., Andrews, Margo, Bartlett, Melissa, Carlson, Nanette, Cox, Emily, Equi, Robert, Gonzalez, Marta, Griffin, Sophia, Hogue, Fran, Kennedy, Lance, Kryuchkov, Lana, Lopez, Carmen, Lopez, Danny, Luevano, Bertha, McKenna, Erin, Patel, Arun, Reed, Brian, Secor, Nyla, Sison, Iris R., Tsai, Tony, Varghis, Nina, Waller, Brooke, Wendel, Robert, Yebra, Reina, Roth, Daniel B., Deinzer, Jane, Fine, Howard, Green, Flory, Green, Stuart, Keyser, Bruce, Leff, Steven, Leviton, Amy, Martir, Amy, Mosenthine, Kristin, Muscle, Starr, Okoren, Linda, Parker, Sandy, Prenner, Jonathan, Price, Nancy, Rogers, Deana, Rosas, Linda, Schlosser, Alex, Studenko, Loretta, Tantum, Thea, Wheatley, Harold, Trese, Michael T., Aaberg, Thomas, Bezaire, Denis, Bridges, Craig, Bryant, Doug, Capone, Antonio, Coleman, Michelle, Consolo, Christina, Cook, Cindy, DuLong, Candice, Garretson, Bruce, Grooten, Tracy, Hammersley, Julie, Hassan, Tarek, Jessick, Heather, Jones, Nanette, Kinsman, Crystal, Krumlauf, Jennifer, Lewis, Sandy, Locke, Heather, Margherio, Alan, Markus, Debra, Marsh, Tanya, Neal, Serena, Noffke, Amy, Oh, Kean, Pence, Clarence, Preston, Lisa, Raphaelian, Paul, Regan, Virginia R., Roberts, Peter, Ruby, Alan, Sarrafizadeh, Ramin, Scherf, Marissa, Scott, Sarita, Sneed, Scott, Staples, Lisa, Terry, Brad, Trese, Matthew T., Videtich, Joan, Williams, George, Zajechowski, Mary, Joseph, Daniel P., Blinder, Kevin, Boyd, Lynda, Buckley, Sarah, Crow, Meaghan, Dinatale, Amanda, Engelbrecht, Nicholas, Forke, Bridget, Gabel, Dana, Grand, Gilbert, Grillion-Cerone, Jennifer, Holekamp, Nancy, Kelly, Charlotte, Nobel, Ginny, Pepple, Kelly, Raeber, Matt, Rao, P. Kumar, Ressel, Tammy, Schremp, Steven, Sgorlon, Merrilee, Shears, Shantia, Thomas, Matthew, Timma, Cathy, Vaughn, Annette, Walters, Carolyn, Weeks, Rhonda, Wehmeier, Jarrod, Wright, Tim, Berinstein, Daniel M., Ayyad, Aida, Barazi, Mohammed K., Bickhart, Erica, Brady, Tracey, Byank, Lisa, Cronise, Alysia, Denny, Vanessa, Dunn, Courtney, Flory, Michael, Frantz, Robert, Garfinkel, Richard A., Gilbert, William, Lai, Michael M., Melamud, Alexander, Newgen, Janine, Newton, Shamekia, Oliver, Debbie, Osman, Michael, Sanders, Reginald, von Fricken, Manfred, Dugel, Pravin, Arenas, Sandra, Balea, Gabe, Bartoli, Dayna, Bucci, John, Cornelius, Jennifer A., Dickens, Scheleen, Doherty, Don, Dunlap, Heather, Goldenberg, David, Jamal, Karim, Jimenez, Norma, Kavanagh, Nicole, Kunimoto, Derek, Martin, John, Miner, Jessica, Mobley, Sarah, Park, Donald, Quinlan, Edward, Sipperley, Jack, Slagle, Carol, Smith, Danielle, Yafchak, Miguelina, Yager, Rohana, Flaxel, Christina J., Bailey, Steven, Francis, Peter, Howell, Chris, Hwang, Thomas, Ira, Shirley, Klein, Michael, Lauer, Andreas, Liesegang, Teresa, Lundquist, Ann, Nolte, Sarah, Nolte, Susan K., Pickell, Scott, Pope, Susan, Rossi, Joseph, Schain, Mitchell, Steinkamp, Peter, Toomey, Maureen D., Vahrenwald, Debora, West, Kelly, Hubbard, Baker, Andelman, Stacey, Bergstrom, Chris, Brower, Judy, Cribbs, Blaine, Curtis, Linda, Dobbs, Jannah, DuBois, Lindreth, Gaultney, Jessica, Gibbs, Deborah, Jordan, Debora, Leef, Donna, Martin, Daniel F., Myles, Robert, Olsen, Timothy, Schwent, Bryan, Srivastava, Sunil, Waldron, Rhonda, Antoszyk, Andrew N., Balasubramaniam, Uma, Brooks, Danielle, Brown, Justin, Browning, David, Clark, Loraine, Ennis, Sarah, Held, Susannah, Helms, Jennifer V., Herby, Jenna, Karow, Angie, Leotaud, Pearl, Massimino, Caterina, McClain, Donna, McOwen, Michael, Mindel, Jennifer, Pereira, Candace, Pierce, Rachel, Powers, Michele, Price, Angela, Rohrer, Jason, Sanders, Jason, Avery, Robert L., Avery, Kelly, Basefsky, Jessica, Beckner, Liz, Castellarin, Alessandro, Couvillion, Stephen, Giust, Jack, Giust, Matthew, Nasir, Maan, Pieramici, Dante, Rabena, Melvin, Risard, Sarah, See, Robert, Smith, Jerry, Wan, Lisha, Bakri, Sophie J., Abu-Yaghi, Nakhleh, Barkmeier, Andrew, Berg, Karin, Burrington, Jean, Edwards, Albert, Goddard, Shannon, Howard, Shannon, Iezzi, Raymond, Lewison, Denise, Link, Thomas, McCannel, Colin A., Overend, Joan, Pach, John, Ruszczyk, Margaret, Shultz, Ryan, Stephan, Cindy, Vogen, Diane, Bradford, Reagan H., Jr., Bergman, Vanessa, Burris, Russ, Butt, Amanda, Daniels, Beth, Dwiggins, Connie, Fransen, Stephen, Guerrero, Tiffany, Haivala, Darin, Harris, Amy, Icks, Sonny, Kingsley, Ronald, Redden, Lena, Richmond, Rob, Ross, Brittany, White, Kammerin, Youngberg, Misty, Topping, Trexler M., Bennett, Steve, Chong, Sandy, Ciotti, Mary, Cleary, Tina, Corey, Emily, Donovan, Dennis, Frederick, Albert, Freese, Lesley, Graham, Margaret, Gud, Natalya, Howard, Taneika, Jones, Mike, Morley, Michael, Moses, Katie, Stone, Jen, Ty, Robin, Wiegand, Torsten, Williams, Lindsey, Winder, Beth, Awh, Carl C., Amonette, Michelle, Arrindell, Everton, Beck, Dena, Busbee, Brandon, Dilback, Amy, Downs, Sara, Guidry, Allison, Gutow, Gary, Hardin, Jackey, Hines, Sarah, Hutchins, Emily, LaCivita, Kim, Lester, Ashley, Malott, Larry, McCain, MaryAnn, Miracle, Jayme, Moffat, Kenneth, Palazzotta, Lacy, Robinson, Kelly, Sonkin, Peter, Travis, Alecia, Wallace, Roy Trent, Winters, Kelly J., Wray, Julia, Harris, April E., Bunnell, Mari, Crooks, Katrina, Fitzgerald, Rebecca, Javid, Cameron, Kew, Corin, Kill, Erica, Kline, Patricia, Kreienkamp, Janet, Martinez, Maricruz, Moore, Roy Ann, Saavedra, Egbert, Taylor, LuAnne, Walsh, Mark, Wilson, Larry, Ciulla, Thomas A., Coyle, Ellen, Harrington, Tonya, Harris, Charlotte, Hood, Cindi, Kerr, Ingrid, Maturi, Raj, Moore, Dawn, Morrow, Stephanie, Savage, Jennifer, Sink, Bethany, Steele, Tom, Thukral, Neelam, Wilburn, Janet, Walker, Joseph P., Banks, Jennifer, Ciampaglia, Debbie, Dyshanowitz, Danielle, Frederick, Jennifer, Ghuman, A. Tom, Grodin, Richard, Kiesel, Cheryl, Knips, Eileen, McCue, Jonathan, Ortiz, Maria, Peters, Crystal, Raskauskas, Paul, Schoeman, Etienne, Sharma, Ashish, Wing, Glenn, Youngblood, Rebecca, Chandra, Suresh R., Altaweel, Michael, Blodi, Barbara, Burke, Kathryn, Dietzman, Kristine A., Gottlieb, Justin, Knutson, Gene, Krolnik, Denise, Nork, T. Michael, Olson, Shelly, Peterson, John, Reed, Sandra, Soderling, Barbara, Somers, Guy, Stevens, Thomas, Wealti, Angela, Bearelly, Srilaxmi, Branchaud, Brenda, Bryant, Joyce W., Crowell, Sara, Fekrat, Sharon, Gammage, Merritt, Harrison, Cheala, Jones, Sarah, McClain, Noreen, McCuen, Brooks, Mruthyunjaya, Prithvi, Queen, Jeanne, Sarin, Neeru, Skalak, Cindy, Skelly, Marriner, Suner, Ivan, Tomany, Ronnie, Welch, Lauren, Park, Susanna S., Cassidy, Allison, Chandra, Karishma, Good, Idalew, Imson, Katrina, Sashi, Kaur, Metzler, Helen, Morse, Lawrence, Redenbo, Ellen, Salvador, Marisa, Telander, David, Thomas, Mark, Wallace, Cindy, Barr, Charles C., Battcher, Amanda, Bottorff, Michelle, Chasteen, Mary, Clark, Kelly, Denning, Diane, Schoen, Debra, Schultz, Amy, Tempel, Evie, Wheeler, Lisa, Whittington, Greg K., Stone, Thomas W., Blevins, Todd, Buck, Michelle, Cruz, Lynn, Heath, Wanda, Holcomb, Diana, Isernhagen, Rick, Kidd, Terri, Kitchens, John, Sears, Cathy, Slade, Ed, Van Arsdall, Jeanne, VanHoose, Brenda, Wolfe, Jenny, Wood, William, Zilis, John, Crooks, Carol, Disney, Larry, Liu, Mimi, Petty, Stephen, Sall, Sandra, Folk, James C., Aly, Tracy, Brotherton, Abby, Critser, Douglas, Hinz, Connie J., Karakas, Stefani, Kirschner, Valerie, Lester, Cheyanne, Montague, Cindy, Russell, Stephen, Stockman, Heather, Taylor, Barbara, Verdick, Randy, Walshire, Jean, Thompson, John T., Connell, Barbara, Constantine, Maryanth, Davis, John L., Jr., Gwen Holsapple, Hunter, Lisa, Lenane, C. Nicki, Mitchell, Robin, Russel, Leslie, Sjaarda, Raymond, Brown, David M., Benz, Matthew, Burns, Llewellyn, Carranza, JoLene G., Fish, Richard, Goates, Debra, Hay, Shayla, Jeffers, Theresa, Kegley, Eric, Kubecka, Dallas, McGilvra, Stacy, Richter, Beau, Sneed, Veronica, Stoever, Cary, Tellez, Isabell, Wong, Tien, Kim, Ivana, Andreoli, Christopher, Barresi, Leslie, Brett, Sarah, Callahan, Charlene, Capaccioli, Karen, Carli, William, Coppola, Matthew, Emmanuel, Nicholas, Evans, Claudia, Fagan, Anna, Grillo, Marcia, Head, John, Kieser, Troy, Lee, Elaine, Lord, Ursula, Miretsky, Edward, Palitsch, Kate, Petrin, Todd, Reader, Liz, Reznichenko, Svetlana, Robertson, Mary, Smith, Justin, Vavvas, Demetrios, Wells, John, Cahill, Cassie, Clark, W. Lloyd, Henry, Kayla, Johnson, David, Miller, Peggy, Oliver, LaDetrick, Spivey, Robbin, Swinford, Tiffany, Taylor, Mallie, Lambert, Michael, Chase, Kris, Fredrickson, Debbie, Khawly, Joseph, Lazarte, Valerie, Lowd, Donald, Miller, Pam, Willis, Arthur, Ferrone, Philip J., Almonte, Miguel, Arnott, Rachel, Aviles, Ingrid, Carbon, Sheri, Chitjian, Michael, DAmore, Kristen, Elliott, Christin, Fastenberg, David, Golub, Barry, Graham, Kenneth, Lavorna, AnnMarie, Murphy, Laura, Palomo, Amanda, Puglisi, Christina, Rhee, David, Romero, Juan, Rosenblatt, Brett, Salcedo, Glenda, Schlameuss, Marianne, Shakin, Eric, Sookhai, Vasanti, Kaiser, Richard, Affel, Elizabeth, Brown, Gary, Centinaro, Christina, Fine, Deborah, Fineman, Mitchell, Formoso, Michele, Garg, Sunir, Grande, Lisa, Herbert, Carolyn, Ho, Allen, Hsu, Jason, Jay, Maryann, Lavetsky, Lisa, Liebenbaum, Elaine, Maguire, Joseph, Monsonego, Julia, O’Connor, Lucia, Pierce, Lisa, Regillo, Carl, Rosario, Maria, Spirn, Marc, Vander, James, Walsh, Jennifer, Davidorf, Frederick H., Barnett, Amanda, Chang, Susie, Christoforidis, John, Elliott, Joy, Justice, Heather, Letson, Alan, McKinney, Kathryne, Perry, Jeri, Salerno, Jill A., Savage, Scott, Shelley, Stephen, Singerman, Lawrence J., Coney, Joseph, DuBois, John, DuBois, Kimberly, Greanoff, Gregg, Himmelman, Dianne, Ilc, Mary, McNamara, Elizabeth, Novak, Michael, Pendergast, Scott, Rath, Susan, Smith-Brewer, Sheila, Tanner, Vivian, Weiss, Diane E., Zegarra, Hernando, Halperin, Lawrence, Aramayo, Patricia, Dhalla, Mandeep, Fernandez, Brian, Fernandez, Cindy, Lopez, Jaclyn, Lopez, Monica, Mariano, Jamie, Murphy, Kellie, Sherley, Clifford, Veksler, Rita, Rahhal, Firas, Babikian, Razmig, Boyer, David, Hami, Sepideh, Kessinger, Jeff, Kurokouchi, Janet, Mukarram, Saba, Pachman, Sarah, Protacio, Eric, Sierra, Julio, Tabandeh, Homayoun, Zamboni, Adam, Elman, Michael, Belz, Jennifer, Butcher, Tammy, Cain, Theresa, Coffey, Teresa, Firestone, Dena, Gore, Nancy, Singletary, Pamela, Sotirakos, Peter, Starr, JoAnn, Meredith, Travis A., Barnhart, Cassandra J., Cantrell, Debra, Esquejo-Leon, RonaLyn, Houghton, Odette, Kaur, Harpreet, NDure, Fatoumatta, Glatzer, Ronald, Joffe, Leonard, Schindler, Reid, Xue, Katie, Hua, Peiying, Maguire, Maureen G., Daniel, Ebenezer, Jaffe, Glenn J., Grunwald, Juan E., and Ying, Gui-shuang

Published
2023
Full Text
View/download PDF

41. Inference of single-cell phylogenies from lineage tracing data using Cassiopeia

Author

Jones, Matthew G, Khodaverdian, Alex, Quinn, Jeffrey J, Chan, Michelle M, Hussmann, Jeffrey A, Wang, Robert, Xu, Chenling, Weissman, Jonathan S, and Yosef, Nir

Subjects
Information and Computing Sciences, Biological Sciences, Genetics, Generic health relevance, Algorithms, CRISPR-Cas Systems, Cell Lineage, Humans, Mutation, Phylogeny, Single-Cell Analysis, scRNA-seq, Single cell, Lineage tracing, CRISPR, Environmental Sciences, Bioinformatics
Abstract

The pairing of CRISPR/Cas9-based gene editing with massively parallel single-cell readouts now enables large-scale lineage tracing. However, the rapid growth in complexity of data from these assays has outpaced our ability to accurately infer phylogenetic relationships. First, we introduce Cassiopeia-a suite of scalable maximum parsimony approaches for tree reconstruction. Second, we provide a simulation framework for evaluating algorithms and exploring lineage tracer design principles. Finally, we generate the most complex experimental lineage tracing dataset to date, 34,557 human cells continuously traced over 15 generations, and use it for benchmarking phylogenetic inference approaches. We show that Cassiopeia outperforms traditional methods by several metrics and under a wide variety of parameter regimes, and provide insight into the principles for the design of improved Cas9-enabled recorders. Together, these should broadly enable large-scale mammalian lineage tracing efforts. Cassiopeia and its benchmarking resources are publicly available at www.github.com/YosefLab/Cassiopeia.

Published
2020

44. STMG: A Machine Learning Microgesture Recognition System for Supporting Thumb-Based VR/AR Input

Author

Kin, Kenrick, primary, Wan, Chengde, additional, Koh, Ken, additional, Marin, Andrei, additional, Camgöz, Necati Cihan, additional, Zhang, Yubo, additional, Cai, Yujun, additional, Kovalev, Fedor, additional, Ben-Zacharia, Moshe, additional, Hoople, Shannon, additional, Nunes-Ueno, Marcos, additional, Sanchez-Rodriguez, Mariel, additional, Bhargava, Ayush, additional, Wang, Robert, additional, Sauser, Eric, additional, and Ma, Shugao, additional

Published
2024
Full Text
View/download PDF

46. MP39-15 USE OF A WORKPLACE-BASED ASSESSMENT TO EVALUATE UROLOGY TRAINEES’ PREPAREDNESS FOR INDEPENDENT PRACTICE

Author

Irani, Sarosh, primary, Khoei, Amelia A., additional, Meyer, Kayla, additional, Daignault-Newton, Stephanie, additional, Badalato, Gina, additional, Breyer, Benjamin, additional, Cox, Lindsey, additional, Hampson, Lindsay A., additional, Jaeger, Christopher, additional, Korets, Ruslan, additional, Mayer, Wesley, additional, Mehta, Akanksha, additional, Taylor, Jennifer, additional, Wang, Robert S., additional, and Kraft, Kate H., additional

Published
2024
Full Text
View/download PDF

47. Pelee: A Real-Time Object Detection System on Mobile Devices

Author

Wang, Robert J., Li, Xiang, and Ling, Charles X.

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

An increasing need of running Convolutional Neural Network (CNN) models on mobile devices with limited computing power and memory resource encourages studies on efficient model design. A number of efficient architectures have been proposed in recent years, for example, MobileNet, ShuffleNet, and MobileNetV2. However, all these models are heavily dependent on depthwise separable convolution which lacks efficient implementation in most deep learning frameworks. In this study, we propose an efficient architecture named PeleeNet, which is built with conventional convolution instead. On ImageNet ILSVRC 2012 dataset, our proposed PeleeNet achieves a higher accuracy and over 1.8 times faster speed than MobileNet and MobileNetV2 on NVIDIA TX2. Meanwhile, PeleeNet is only 66% of the model size of MobileNet. We then propose a real-time object detection system by combining PeleeNet with Single Shot MultiBox Detector (SSD) method and optimizing the architecture for fast speed. Our proposed detection system2, named Pelee, achieves 76.4% mAP (mean average precision) on PASCAL VOC2007 and 22.4 mAP on MS COCO dataset at the speed of 23.6 FPS on iPhone 8 and 125 FPS on NVIDIA TX2. The result on COCO outperforms YOLOv2 in consideration of a higher precision, 13.6 times lower computational cost and 11.3 times smaller model size., Comment: Accepted to NeurIPS 2018

Published
2018

48. Going Deeper in Spiking Neural Networks: VGG and Residual Architectures

Author

Sengupta, Abhronil, Ye, Yuting, Wang, Robert, Liu, Chiao, and Roy, Kaushik

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Over the past few years, Spiking Neural Networks (SNNs) have become popular as a possible pathway to enable low-power event-driven neuromorphic hardware. However, their application in machine learning have largely been limited to very shallow neural network architectures for simple problems. In this paper, we propose a novel algorithmic technique for generating an SNN with a deep architecture, and demonstrate its effectiveness on complex visual recognition problems such as CIFAR-10 and ImageNet. Our technique applies to both VGG and Residual network architectures, with significantly better accuracy than the state-of-the-art. Finally, we present analysis of the sparse event-driven computations to demonstrate reduced hardware overhead when operating in the spiking domain.

Published
2018

Catalog

Books, media, physical & digital resources
See catalog results