Search

Semi-supervised 3D object detection is a common strategy employed to circumvent the challenge of manually labeling large-scale autonomous driving perception datasets. Pseudo-labeling approaches to semi-supervised learning adopt a teacher-student framework in which machine-generated pseudo-labels on a large unlabeled dataset are used in combination with a small manually-labeled dataset for training. In this work, we address the problem of improving pseudo-label quality through leveraging long-term temporal information captured in driving scenes. More specifically, we leverage pre-trained motion-forecasting models to generate object trajectories on pseudo-labeled data to further enhance the student model training. Our approach improves pseudo-label quality in two distinct manners: first, we suppress false positive pseudo-labels through establishing consistency across multiple frames of motion forecasting outputs. Second, we compensate for false negative detections by directly inserting predicted object tracks into the pseudo-labeled scene. Experiments on the nuScenes dataset demonstrate the effectiveness of our approach, improving the performance of standard semi-supervised approaches in a variety of settings.

Recent 3D novel view synthesis (NVS) methods are limited to single-object-centric scenes generated from new viewpoints and struggle with complex environments. They often require extensive 3D data for training, lacking generalization beyond training distribution. Conversely, 3D-free methods can generate text-controlled views of complex, in-the-wild scenes using a pretrained stable diffusion model without tedious fine-tuning, but lack camera control. In this paper, we introduce HawkI++, a method capable of generating camera-controlled viewpoints from a single input image. HawkI++ excels in handling complex and diverse scenes without additional 3D data or extensive training. It leverages widely available pretrained NVS models for weak guidance, integrating this knowledge into a 3D-free view synthesis approach to achieve the desired results efficiently. Our experimental results demonstrate that HawkI++ outperforms existing models in both qualitative and quantitative evaluations, providing high-fidelity and consistent novel view synthesis at desired camera angles across a wide variety of scenes., Comment: 6 pages, 7 figures

Data for training learning-enabled self-driving cars in the physical world are typically collected in a safe, normal environment. Such data distribution often engenders a strong bias towards safe driving, making self-driving cars unprepared when encountering adversarial scenarios like unexpected accidents. Due to a dearth of such adverse data that is unrealistic for drivers to collect, autonomous vehicles can perform poorly when experiencing such rare events. This work addresses much-needed research by having participants drive a VR vehicle simulator going through simulated traffic with various types of accidental scenarios. It aims to understand human responses and behaviors in simulated accidents, contributing to our understanding of driving dynamics and safety. The simulation framework adopts a robust traffic simulation and is rendered using the Unity Game Engine. Furthermore, the simulation framework is built with portable, light-weight immersive driving simulator hardware, lowering the resource barrier for studies in autonomous driving research. Keywords: Rare Events, Traffic Simulation, Autonomous Driving, Virtual Reality, User Studies

Vector fields are widely used to represent and model flows for many science and engineering applications. This paper introduces a novel neural network architecture for learning tangent vector fields that are intrinsically defined on manifold surfaces embedded in 3D. Previous approaches to learning vector fields on surfaces treat vectors as multi-dimensional scalar fields, using traditional scalar-valued architectures to process channels individually, thus fail to preserve fundamental intrinsic properties of the vector field. The core idea of this work is to introduce a trainable vector heat diffusion module to spatially propagate vector-valued feature data across the surface, which we incorporate into our proposed architecture that consists of vector-valued neurons. Our architecture is invariant to rigid motion of the input, isometric deformation, and choice of local tangent bases, and is robust to discretizations of the surface. We evaluate our Vector Heat Network on triangle meshes, and empirically validate its invariant properties. We also demonstrate the effectiveness of our method on the useful industrial application of quadrilateral mesh generation.

In trajectory forecasting tasks for traffic, future output trajectories can be computed by advancing the ego vehicle's state with predicted actions according to a kinematics model. By unrolling predicted trajectories via time integration and models of kinematic dynamics, predicted trajectories should not only be kinematically feasible but also relate uncertainty from one timestep to the next. While current works in probabilistic prediction do incorporate kinematic priors for mean trajectory prediction, _variance_ is often left as a learnable parameter, despite uncertainty in one time step being inextricably tied to uncertainty in the previous time step. In this paper, we show simple and differentiable analytical approximations describing the relationship between variance at one timestep and that at the next with the kinematic bicycle model. In our results, we find that encoding the relationship between variance across timesteps works especially well in unoptimal settings, such as with small or noisy datasets. We observe up to a 50% performance boost in partial dataset settings and up to an 8% performance boost in large-scale learning compared to previous kinematic prediction methods on SOTA trajectory forecasting architectures out-of-the-box, with no fine-tuning., Comment: 8 pages

Segmentation is an integral module in many visual computing applications such as virtual try-on, medical imaging, autonomous driving, and agricultural automation. These applications often involve either widespread consumer use or highly variable environments, both of which can degrade the quality of visual sensor data, whether from a common mobile phone or an expensive satellite imaging camera. In addition to external noises like user difference or weather conditions, internal noises such as variations in camera quality or lens distortion can affect the performance of segmentation models during both development and deployment. In this work, we present an efficient, adaptable, and gradient-free method to enhance the robustness of learning-based segmentation models across training. First, we introduce a novel adaptive sensitivity analysis (ASA) using Kernel Inception Distance (KID) on basis perturbations to benchmark perturbation sensitivity of pre-trained segmentation models. Then, we model the sensitivity curve using the adaptive SA and sample perturbation hyperparameter values accordingly. Finally, we conduct adversarial training with the selected perturbation values and dynamically re-evaluate robustness during online training. Our method, implemented end-to-end with minimal fine-tuning required, consistently outperforms state-of-the-art data augmentation techniques for segmentation. It shows significant improvement in both clean data evaluation and real-world adverse scenario evaluation across various segmentation datasets used in visual computing and computer graphics applications., Comment: 10 pages

We present a differentiable representation, DMesh, for general 3D triangular meshes. DMesh considers both the geometry and connectivity information of a mesh. In our design, we first get a set of convex tetrahedra that compactly tessellates the domain based on Weighted Delaunay Triangulation (WDT), and select triangular faces on the tetrahedra to define the final mesh. We formulate probability of faces to exist on the actual surface in a differentiable manner based on the WDT. This enables DMesh to represent meshes of various topology in a differentiable way, and allows us to reconstruct the mesh under various observations, such as point cloud and multi-view images using gradient-based optimization. The source code and full paper is available at: https://sonsang.github.io/dmesh-project., Comment: 35 pages, 22 figures. Updated with more analysis and experimental results

We introduce a novel policy learning method that integrates analytical gradients from differentiable environments with the Proximal Policy Optimization (PPO) algorithm. To incorporate analytical gradients into the PPO framework, we introduce the concept of an {\alpha}-policy that stands as a locally superior policy. By adaptively modifying the {\alpha} value, we can effectively manage the influence of analytical policy gradients during learning. To this end, we suggest metrics for assessing the variance and bias of analytical gradients, reducing dependence on these gradients when high variance or bias is detected. Our proposed approach outperforms baseline algorithms in various scenarios, such as function optimization, physics simulations, and traffic control environments. Our code can be found online: https://github.com/SonSang/gippo., Comment: 27 pages, NeurIPS 2023 Conference

Collaborative decision-making is an essential capability for multi-robot systems, such as connected vehicles, to collaboratively control autonomous vehicles in accident-prone scenarios. Under limited communication bandwidth, capturing comprehensive situational awareness by integrating connected agents' observation is very challenging. In this paper, we propose a novel collaborative decision-making method that efficiently and effectively integrates collaborators' representations to control the ego vehicle in accident-prone scenarios. Our approach formulates collaborative decision-making as a classification problem. We first represent sequences of raw observations as spatiotemporal graphs, which significantly reduce the package size to share among connected vehicles. Then we design a novel spatiotemporal graph neural network based on heterogeneous graph learning, which analyzes spatial and temporal connections of objects in a unified way for collaborative decision-making. We evaluate our approach using a high-fidelity simulator that considers realistic traffic, communication bandwidth, and vehicle sensing among connected autonomous vehicles. The experimental results show that our representation achieves over 100x reduction in the shared data size that meets the requirements of communication bandwidth for connected autonomous driving. In addition, our approach achieves over 30% improvements in driving safety.

We present a novel learning-based trajectory generation algorithm for outdoor robot navigation. Our goal is to compute collision-free paths that also satisfy the environment-specific traversability constraints. Our approach is designed for global planning using limited onboard robot perception in mapless environments while ensuring comprehensive coverage of all traversable directions. Our formulation uses a Conditional Variational Autoencoder (CVAE) generative model that is enhanced with traversability constraints and an optimization formulation used for the coverage. We highlight the benefits of our approach over state-of-the-art trajectory generation approaches and demonstrate its performance in challenging and large outdoor environments, including around buildings, across intersections, along trails, and off-road terrain, using a Clearpath Husky and a Boston Dynamics Spot robot. In practice, our approach results in a 6% improvement in coverage of traversable areas and an 89% reduction in trajectory portions residing in non-traversable regions. Our video is here: https://youtu.be/3eJ2soAzXnU, Comment: 9

Embedding polygonal mesh assets within photorealistic Neural Radience Fields (NeRF) volumes, such that they can be rendered and their dynamics simulated in a physically consistent manner with the NeRF, is under-explored from the system perspective of integrating NeRF into the traditional graphics pipeline. This paper designs a two-way coupling between mesh and NeRF during rendering and simulation. We first review the light transport equations for both mesh and NeRF, then distill them into an efficient algorithm for updating radiance and throughput along a cast ray with an arbitrary number of bounces. To resolve the discrepancy between the linear color space that the path tracer assumes and the sRGB color space that standard NeRF uses, we train NeRF with High Dynamic Range (HDR) images. We also present a strategy to estimate light sources and cast shadows on the NeRF. Finally, we consider how the hybrid surface-volumetric formulation can be efficiently integrated with a high-performance physics simulator that supports cloth, rigid and soft bodies. The full rendering and simulation system can be run on a GPU at interactive rates. We show that a hybrid system approach outperforms alternatives in visual realism for mesh insertion, because it allows realistic light transport from volumetric NeRF media onto surfaces, which affects the appearance of reflective/refractive surfaces and illumination of diffuse surfaces informed by the dynamic scene., Comment: ICCV 2023

Repetitive counting (RepCount) is critical in various applications, such as fitness tracking and rehabilitation. Previous methods have relied on the estimation of red-green-and-blue (RGB) frames and body pose landmarks to identify the number of action repetitions, but these methods suffer from a number of issues, including the inability to stably handle changes in camera viewpoints, over-counting, under-counting, difficulty in distinguishing between sub-actions, inaccuracy in recognizing salient poses, etc. In this paper, based on the work done by [1], we integrate joint angles with body pose landmarks to address these challenges and achieve better results than the state-of-the-art RepCount methods, with a Mean Absolute Error (MAE) of 0.211 and an Off-By-One (OBO) counting accuracy of 0.599 on the RepCount data set [2]. Comprehensive experimental results demonstrate the effectiveness and robustness of our method., Comment: 7 pages, 9 figures

A thermal simulation methodology derived from the proper orthogonal decomposition (POD) and the Galerkin projection (GP), hereafter referred to as PODTherm-GP, is evaluated in terms of its efficiency and accuracy in a multi-core CPU. The GP projects the heat transfer equation onto a mathematical space whose basis functions are generated from thermal data enabled by the POD learning algorithm. The thermal solution data are collected from FEniCS using the finite element method (FEM) accounting for appropriate parametric variations. The GP incorporates physical principles of heat transfer in the methodology to reach high accuracy and efficiency. The dynamic power map for the CPU in FEM thermal simulation is generated from gem5 and McPACT, together with the SPLASH-2 benchmarks as the simulation workload. It is shown that PODTherm-GP offers an accurate thermal prediction of the CPU with a resolution as fine as the FEM. It is also demonstrated that PODTherm-GP is capable of predicting the dynamic thermal profile of the chip with a good accuracy beyond the training conditions. Additionally, the approach offers a reduction in degrees of freedom by more than 5 orders of magnitude and a speedup of 4 orders, compared to the FEM.

Autonomous driving research currently faces data sparsity in representation of risky scenarios. Such data is both difficult to obtain ethically in the real world, and unreliable to obtain via simulation. Recent advances in virtual reality (VR) driving simulators lower barriers to tackling this problem in simulation. We propose the first data collection framework for risky scenario driving data from real humans using VR, as well as accompanying numerical driving personality characterizations. We validate the resulting dataset with statistical analyses and model driving behavior with an eight-factor personality vector based on the Multi-dimensional Driving Style Inventory (MDSI). Our method, dataset, and analyses show that realistic driving personalities can be modeled without deep learning or large datasets to complement autonomous driving research.

Haptic feedback is an important component of creating an immersive mixed reality experience. Traditionally, haptic forces are rendered in response to the user's interactions with the virtual environment. In this work, we explore the idea of rendering haptic forces in a proactive manner, with the explicit intention to influence the user's behavior through compelling haptic forces. To this end, we present a framework for active haptic guidance in mixed reality, using one or more robotic haptic proxies to influence user behavior and deliver a safer and more immersive virtual experience. We provide details on common challenges that need to be overcome when implementing active haptic guidance, and discuss example applications that show how active haptic guidance can be used to influence the user's behavior. Finally, we apply active haptic guidance to a virtual reality navigation problem, and conduct a user study that demonstrates how active haptic guidance creates a safer and more immersive experience for users.

We present a method for learning 3D geometry and physics parameters of a dynamic scene from only a monocular RGB video input. To decouple the learning of underlying scene geometry from dynamic motion, we represent the scene as a time-invariant signed distance function (SDF) which serves as a reference frame, along with a time-conditioned deformation field. We further bridge this neural geometry representation with a differentiable physics simulator by designing a two-way conversion between the neural field and its corresponding hexahedral mesh, enabling us to estimate physics parameters from the source video by minimizing a cycle consistency loss. Our method also allows a user to interactively edit 3D objects from the source video by modifying the recovered hexahedral mesh, and propagating the operation back to the neural field representation. Experiments show that our method achieves superior mesh and video reconstruction of dynamic scenes compared to competing Neural Field approaches, and we provide extensive examples which demonstrate its ability to extract useful 3D representations from videos captured with consumer-grade cameras., Comment: NeurIPS 2022

We introduce a novel differentiable hybrid traffic simulator, which simulates traffic using a hybrid model of both macroscopic and microscopic models and can be directly integrated into a neural network for traffic control and flow optimization. This is the first differentiable traffic simulator for macroscopic and hybrid models that can compute gradients for traffic states across time steps and inhomogeneous lanes. To compute the gradient flow between two types of traffic models in a hybrid framework, we present a novel intermediate conversion component that bridges the lanes in a differentiable manner as well. We also show that we can use analytical gradients to accelerate the overall process and enhance scalability. Thanks to these gradients, our simulator can provide more efficient and scalable solutions for complex learning and control problems posed in traffic engineering than other existing algorithms. Refer to https://sites.google.com/umd.edu/diff-hybrid-traffic-sim for our project., Comment: 13 pages, Siggraph Asia 2022 Journal Paper

We present a Body Measurement network (BMnet) for estimating 3D anthropomorphic measurements of the human body shape from silhouette images. Training of BMnet is performed on data from real human subjects, and augmented with a novel adversarial body simulator (ABS) that finds and synthesizes challenging body shapes. ABS is based on the skinned multiperson linear (SMPL) body model, and aims to maximize BMnet measurement prediction error with respect to latent SMPL shape parameters. ABS is fully differentiable with respect to these parameters, and trained end-to-end via backpropagation with BMnet in the loop. Experiments show that ABS effectively discovers adversarial examples, such as bodies with extreme body mass indices (BMI), consistent with the rarity of extreme-BMI bodies in BMnet's training set. Thus ABS is able to reveal gaps in training data and potential failures in predicting under-represented body shapes. Results show that training BMnet with ABS improves measurement prediction accuracy on real bodies by up to 10%, when compared to no augmentation or random body shape sampling. Furthermore, our method significantly outperforms SOTA measurement estimation methods by as much as 3x. Finally, we release BodyM, the first challenging, large-scale dataset of photo silhouettes and body measurements of real human subjects, to further promote research in this area. Project website: https://adversarialbodysim.github.io, Comment: Published at the International Conference on 3D Vision (3DV) 2022

While there have been advancements in autonomous driving control and traffic simulation, there have been little to no works exploring their unification with deep learning. Works in both areas seem to focus on entirely different exclusive problems, yet traffic and driving are inherently related in the real world. In this paper, we present Traffic-Aware Autonomous Driving (TrAAD), a generalizable distillation-style method for traffic-informed imitation learning that directly optimizes for faster traffic flow and lower energy consumption. TrAAD focuses on the supervision of speed control in imitation learning systems, as most driving research focuses on perception and steering. Moreover, our method addresses the lack of co-simulation between traffic and driving simulators and provides a basis for directly involving traffic simulation with autonomous driving in future work. Our results show that, with information from traffic simulation involved in the supervision of imitation learning methods, an autonomous vehicle can learn how to accelerate in a fashion that is beneficial for traffic flow and overall energy consumption for all nearby vehicles.

Leveraging multi-modal fusion, especially between camera and LiDAR, has become essential for building accurate and robust 3D object detection systems for autonomous vehicles. Until recently, point decorating approaches, in which point clouds are augmented with camera features, have been the dominant approach in the field. However, these approaches fail to utilize the higher resolution images from cameras. Recent works projecting camera features to the bird's-eye-view (BEV) space for fusion have also been proposed, however they require projecting millions of pixels, most of which only contain background information. In this work, we propose a novel approach Center Feature Fusion (CFF), in which we leverage center-based detection networks in both the camera and LiDAR streams to identify relevant object locations. We then use the center-based detection to identify the locations of pixel features relevant to object locations, a small fraction of the total number in the image. These are then projected and fused in the BEV frame. On the nuScenes dataset, we outperform the LiDAR-only baseline by 4.9% mAP while fusing up to 100x fewer features than other fusion methods., Comment: Accepted by ICRA 2023

We present a method for differentiable simulation of soft articulated bodies. Our work enables the integration of differentiable physical dynamics into gradient-based pipelines. We develop a top-down matrix assembly algorithm within Projective Dynamics and derive a generalized dry friction model for soft continuum using a new matrix splitting strategy. We derive a differentiable control framework for soft articulated bodies driven by muscles, joint torques, or pneumatic tubes. The experiments demonstrate that our designs make soft body simulation more stable and realistic compared to other frameworks. Our method accelerates the solution of system identification problems by more than an order of magnitude, and enables efficient gradient-based learning of motion control with soft robots., Comment: NeurIPS 2021

Human Activity Recognition (HAR) using wearable devices such as smart watches embedded with Inertial Measurement Unit (IMU) sensors has various applications relevant to our daily life, such as workout tracking and health monitoring. In this paper, we propose a novel attention-based approach to human activity recognition using multiple IMU sensors worn at different body locations. Firstly, a sensor-wise feature extraction module is designed to extract the most discriminative features from individual sensors with Convolutional Neural Networks (CNNs). Secondly, an attention-based fusion mechanism is developed to learn the importance of sensors at different body locations and to generate an attentive feature representation. Finally, an inter-sensor feature extraction module is applied to learn the inter-sensor correlations, which are connected to a classifier to output the predicted classes of activities. The proposed approach is evaluated using five public datasets and it outperforms state-of-the-art methods on a wide variety of activity categories.

A thermal simulation methodology is developed for interconnects enabled by a data-driven learning algorithm accounting for variations of material properties, heat sources and boundary conditions (BCs). The methodology is based on the concepts of model order reduction and domain decomposition to construct a multi-block approach. A generic block model is built to represent a group of interconnect blocks that are used to wire standard cells in the integrated circuits (ICs). The blocks in this group possess identical geometry with various metal/via routings. The data-driven model reduction method is thus applied to learn material property variations induced by different metal/via routings in the blocks, in addition to the variations of heat sources and BCs. The approach is investigated in two very different settings. It is first applied to thermal simulation of a single interconnect block with similar BCs to those in the training of the generic block. It is then implemented in multi-block thermal simulation of a FinFET IC, where the interconnect structure is partitioned into several blocks each modeled by the generic block model. Accuracy of the generic block model is examined in terms of the metal/via routings, BCs and thermal discontinuities at the block interfaces., Comment: 23 pages, 15 figures

Face aging techniques have used generative adversarial networks (GANs) and style transfer learning to transform one's appearance to look younger/older. Identity is maintained by conditioning these generative networks on a learned vector representation of the source content. In this work, we apply a similar approach to age a speaker's voice, referred to as voice aging. We first analyze the classification of a speaker's age by training a convolutional neural network (CNN) on the speaker's voice and face data from Common Voice and VoxCeleb datasets. We generate aged voices from style transfer to transform an input spectrogram to various ages and demonstrate our method on a mobile app.

Reflective and textureless surfaces such as windows, mirrors, and walls can be a challenge for object and scene reconstruction. These surfaces are often poorly reconstructed and filled with depth discontinuities and holes, making it difficult to cohesively reconstruct scenes that contain these planar discontinuities. We propose Echoreconstruction, an audio-visual method that uses the reflections of sound to aid in geometry and audio reconstruction for virtual conferencing, teleimmersion, and other AR/VR experience. The mobile phone prototype emits pulsed audio, while recording video for RGB-based 3D reconstruction and audio-visual classification. Reflected sound and images from the video are input into our audio (EchoCNN-A) and audio-visual (EchoCNN-AV) convolutional neural networks for surface and sound source detection, depth estimation, and material classification. The inferences from these classifications enhance scene 3D reconstructions containing open spaces and reflective surfaces by depth filtering, inpainting, and placement of unmixed sound sources in the scene. Our prototype, VR demo, and experimental results from real-world and virtual scenes with challenging surfaces and sound indicate high success rates on classification of material, depth estimation, and closed/open surfaces, leading to considerable visual and audio improvement in 3D scenes (see Figure 1).

3D object reconstructions of transparent and concave structured objects, with inferred material properties, remains an open research problem for robot navigation in unstructured environments. In this paper, we propose a multimodal single- and multi-frame neural network for 3D reconstructions using audio-visual inputs. Our trained reconstruction LSTM autoencoder 3D-MOV accepts multiple inputs to account for a variety of surface types and views. Our neural network produces high-quality 3D reconstructions using voxel representation. Based on Intersection-over-Union (IoU), we evaluate against other baseline methods using synthetic audio-visual datasets ShapeNet and Sound20K with impact sounds and bounding box annotations. To the best of our knowledge, our single- and multi-frame model is the first audio-visual reconstruction neural network for 3D geometry and material representation.

We present a method for efficient differentiable simulation of articulated bodies. This enables integration of articulated body dynamics into deep learning frameworks, and gradient-based optimization of neural networks that operate on articulated bodies. We derive the gradients of the forward dynamics using spatial algebra and the adjoint method. Our approach is an order of magnitude faster than autodiff tools. By only saving the initial states throughout the simulation process, our method reduces memory requirements by two orders of magnitude. We demonstrate the utility of efficient differentiable dynamics for articulated bodies in a variety of applications. We show that reinforcement learning with articulated systems can be accelerated using gradients provided by our method. In applications to control and inverse problems, gradient-based optimization enabled by our work accelerates convergence by more than an order of magnitude., Comment: ICML 2021