Search

We present Follow-Your-Emoji, a diffusion-based framework for portrait animation, which animates a reference portrait with target landmark sequences. The main challenge of portrait animation is to preserve the identity of the reference portrait and transfer the target expression to this portrait while maintaining temporal consistency and fidelity. To address these challenges, Follow-Your-Emoji equipped the powerful Stable Diffusion model with two well-designed technologies. Specifically, we first adopt a new explicit motion signal, namely expression-aware landmark, to guide the animation process. We discover this landmark can not only ensure the accurate motion alignment between the reference portrait and target motion during inference but also increase the ability to portray exaggerated expressions (i.e., large pupil movements) and avoid identity leakage. Then, we propose a facial fine-grained loss to improve the model's ability of subtle expression perception and reference portrait appearance reconstruction by using both expression and facial masks. Accordingly, our method demonstrates significant performance in controlling the expression of freestyle portraits, including real humans, cartoons, sculptures, and even animals. By leveraging a simple and effective progressive generation strategy, we extend our model to stable long-term animation, thus increasing its potential application value. To address the lack of a benchmark for this field, we introduce EmojiBench, a comprehensive benchmark comprising diverse portrait images, driving videos, and landmarks. We show extensive evaluations on EmojiBench to verify the superiority of Follow-Your-Emoji., Comment: Project Page: https://follow-your-emoji.github.io/

Several open-source systems, such as Flower and NVIDIA FLARE, have been developed in recent years while focusing on different aspects of federated learning (FL). Flower is dedicated to implementing a cohesive approach to FL, analytics, and evaluation. Over time, Flower has cultivated extensive strategies and algorithms tailored for FL application development, fostering a vibrant FL community in research and industry. Conversely, FLARE has prioritized the creation of an enterprise-ready, resilient runtime environment explicitly designed for FL applications in production environments. In this paper, we describe our initial integration of both frameworks and show how they can work together to supercharge the FL ecosystem as a whole. Through the seamless integration of Flower and FLARE, applications crafted within the Flower framework can effortlessly operate within the FLARE runtime environment without necessitating any modifications. This initial integration streamlines the process, eliminating complexities and ensuring smooth interoperability between the two platforms, thus enhancing the overall efficiency and accessibility of FL applications.

Federated learning (FL) is a distributed learning paradigm that facilitates collaborative training of a shared global model across devices while keeping data localized. The deployment of FL in numerous real-world applications faces delays, primarily due to the prevalent reliance on supervised tasks. Generating detailed labels at edge devices, if feasible, is demanding, given resource constraints and the imperative for continuous data updates. In addressing these challenges, solutions such as federated semi-supervised learning (FSSL), which relies on unlabeled clients' data and a limited amount of labeled data on the server, become pivotal. In this paper, we propose FedAnchor, an innovative FSSL method that introduces a unique double-head structure, called anchor head, paired with the classification head trained exclusively on labeled anchor data on the server. The anchor head is empowered with a newly designed label contrastive loss based on the cosine similarity metric. Our approach mitigates the confirmation bias and overfitting issues associated with pseudo-labeling techniques based on high-confidence model prediction samples. Extensive experiments on CIFAR10/100 and SVHN datasets demonstrate that our method outperforms the state-of-the-art method by a significant margin in terms of convergence rate and model accuracy.

Recently, the growing interest in wearable technology for personal healthcare and smart VR/AR applications newly imposed a need for development of facile fabrication method. Regarding the issue, laser has long been proposing original answers to such challenging technological demands with its remote, sterile, rapid, and site-selective processing characteristics for arbitrary materials. In this review, recent developments in relevant laser processes are summarized in two separate categories. Firstly, transformative approaches represented by laser-induced graphene (LIG) are introduced. Apart from design optimization and alteration of native substrate, latest advancements in the transformative approach now enable not only more complex material compositions but also multilayer device configurations by simultaneous transformation of heterogeneous precursor or sequential addition of functional layers coupled with other electronic elements. Besides, more conventional laser techniques such as ablation, sintering and synthesis are still accessible for enhancing the functionality of the entire system through expansion of applicable materials and adoption of new mechanisms. Various wearable device components developed through the corresponding laser processes are then organized with emphasis on chemical/physical sensors and energy devices. At the same time, special attention is given to the applications utilizing multiple laser sources or multiple laser processes, which pave the way towards all-laser fabrication of wearable devices.

Most work in privacy-preserving federated learning (FL) has focused on horizontally partitioned datasets where clients hold the same features and train complete client-level models independently. However, individual data points are often scattered across different institutions, known as clients, in vertical FL (VFL) settings. Addressing this category of FL necessitates the exchange of intermediate outputs and gradients among participants, resulting in potential privacy leakage risks and slow convergence rates. Additionally, in many real-world scenarios, VFL training also faces the acute issue of client stragglers and drop-outs, a serious challenge that can significantly hinder the training process but has been largely overlooked in existing studies. In this work, we present vFedSec, a first dropout-tolerant VFL protocol, which can support the most generalized vertical framework. It achieves secure and efficient model training by using an innovative Secure Layer alongside an embedding-padding technique. We provide theoretical proof that our design attains enhanced security while maintaining training performance. Empirical results from extensive experiments also demonstrate vFedSec is robust to client dropout and provides secure training with negligible computation and communication overhead. Compared to widely adopted homomorphic encryption (HE) methods, our approach achieves a remarkable > 690x speedup and reduces communication costs significantly by > 9.6x., Comment: Generalised extension from our previous work: arXiv:2305.11236

The majority of work in privacy-preserving federated learning (FL) has been focusing on horizontally partitioned datasets where clients share the same sets of features and can train complete models independently. However, in many interesting problems, such as financial fraud detection and disease detection, individual data points are scattered across different clients/organizations in vertical federated learning. Solutions for this type of FL require the exchange of gradients between participants and rarely consider privacy and security concerns, posing a potential risk of privacy leakage. In this work, we present a novel design for training vertical FL securely and efficiently using state-of-the-art security modules for secure aggregation. We demonstrate empirically that our method does not impact training performance whilst obtaining 9.1e2 ~3.8e4 speedup compared to homomorphic encryption (HE)., Comment: Federated Learning Systems (FLSys) Workshop @ MLSys 2023

We present a pipeline of Image to Vector (Img2Vec) for masked image modeling (MIM) with deep features. To study which type of deep features is appropriate for MIM as a learning target, we propose a simple MIM framework with serials of well-trained self-supervised models to convert an Image to a feature Vector as the learning target of MIM, where the feature extractor is also known as a teacher model. Surprisingly, we empirically find that an MIM model benefits more from image features generated by some lighter models (e.g., ResNet-50, 26M) than from those by a cumbersome teacher like Transformer-based models (e.g., ViT-Large, 307M). To analyze this remarkable phenomenon, we devise a novel attribute, token diversity, to evaluate the characteristics of generated features from different models. Token diversity measures the feature dissimilarity among different tokens. Through extensive experiments and visualizations, we hypothesize that beyond the acknowledgment that a large model can improve MIM, a high token-diversity of a teacher model is also crucial. Based on the above discussion, Img2Vec adopts a teacher model with high token-diversity to generate image features. Img2Vec pre-trained on ImageNet unlabeled data with ViT-B yields 85.1\% top-1 accuracy on fine-tuning. Moreover, we scale up Img2Vec on larger models, ViT-L and ViT-H, and get $86.7\%$ and $87.5\%$ accuracy respectively. It also achieves state-of-the-art results on other downstream tasks, e.g., 51.8\% mAP on COCO and 50.7\% mIoU on ADE20K. Img2Vec is a simple yet effective framework tailored to deep feature MIM learning, accomplishing superb comprehensive performance on representative vision tasks.

Considering the user mobility and unpredictable mobile edge computing (MEC) environments, this paper studies the intelligent task offloading problem in unmanned aerial vehicle (UAV)-enabled MEC with the assistance of digital twin (DT). We aim at minimizing the energy consumption of the entire MEC system by jointly optimizing mobile terminal users (MTUs) association, UAV trajectory, transmission power distribution and computation capacity allocation while respecting the constraints of mission maximum processing delays. Specifically, double deep Q-network (DDQN) algorithm stemming from deep reinforcement learning is first proposed to effectively solve the problem of MTUs association and UAV trajectory. Then, the closed-form expression is employed to handle the problem of transmission power distribution and the computation capacity allocation problem is further addressed via an iterative algorithm. Numerical results show that our proposed scheme is able to converge and significantly reduce the total energy consumption of the MEC system compared to the benchmark schemes., Comment: 14 pages, 11 figures

Deep learning has been used in a wide range of areas and made a huge breakthrough. With the ever-increasing model size and train-ing data volume, distributed deep learning emerges which utilizes a cluster to train a model in parallel. Unfortunately, the performance is often far from linear speedup due to the communication overhead between cluster nodes. To address this challenge, this paper designs and implements Libra, a network aggregator, that utilizes in-network computation to optimize the communication for distributed DL training in two aspects: 1) reduce active connections and 2) aggregate exchanged network packets. We implemented our Libra on Intel Tofino switches, customized a lightweight host stack and integrated it into an open-source training framework PS-lite. The experimental result shows that our Libra can achieve 1.5~4 times speedup., Comment: 13 pages, 18 figures

The MQXFA Quadrupole magnets will be installed in High Luminosity LHC to form the Q1 and Q3 inner triplet optical elements in front of the interaction points 1 (ATLAS) and 5 (CMS). A pair of MQXFA units is assembled in a stainless steel helium vessel, including the end domes, to make the Q1 Cold Mass or the Q3 Cold Mass. The US HL LHC Accelerator Upgrade Project* is responsible for the design, manufacturing and test of the Q1/Q3 Cold Masses and the complete MQXFA magnets. CERN provides the cryostat components and is responsible for integration and installation in HL LHC. The MQXFA quadrupoles have 150 mm aperture, 4.2 m magnetic length, nominal gradient of 132.2 T/m, and coil peak field of 11.3 T. They use Nb_3Sn conductor and a support structure made of segmented aluminum shells pre-loaded by using bladders and keys. This report presents the final design of the MQXFA quadrupole magnets. *Supported by the U.S. Department of Energy, Office of Science, Office of High Energy Physics, Comment: 120 pages

In-network computation has been widely used to accelerate data-intensive distributed applications. Some computational tasks, traditional performed on servers, are offloaded to the network (i.e. programmable switches). However, the computational capacity of programmable switches is limited to simple integer arithmetic operations while many of applications require on-the-fly floating-point operations. To address this issue, prior approaches either adopt a float-to-integer method or directly offload computational tasks to the local CPUs of switches, incurring accuracy loss and delayed processing. To this end, we propose NetFC, a table-lookup method to achieve on-the-fly in-network floating-point arithmetic operations nearly without accuracy loss. NetFC adopts a divide-and-conquer mechanism that converts the original huge table into several much small tables together with some integer operations. NetFC adopts a scaling-factor mechanism for computational accuracy improvement, and a prefix-based lossless table compression method to reduce the memory overhead. We use different types of datasets to evaluate NetFC. The experimental results show that the average accuracy of NetFC can be as high as up to 99.94% at worst with only 448KB memory consumption. Furthermore, we integrate NetFC into Sonata for detecting Slowloris attack, yielding significant decrease of detection delay., Comment: 10 pages body, 11 pages total

It is well known that adversarial attacks can fool deep neural networks with imperceptible perturbations. Although adversarial training significantly improves model robustness, failure cases of defense still broadly exist. In this work, we find that the adversarial attacks can also be vulnerable to small perturbations. Namely, on adversarially-trained models, perturbing adversarial examples with a small random noise may invalidate their misled predictions. After carefully examining state-of-the-art attacks of various kinds, we find that all these attacks have this deficiency to different extents. Enlightened by this finding, we propose to counter attacks by crafting more effective defensive perturbations. Our defensive perturbations leverage the advantage that adversarial training endows the ground-truth class with smaller local Lipschitzness. By simultaneously attacking all the classes, the misled predictions with larger Lipschitzness can be flipped into correct ones. We verify our defensive perturbation with both empirical experiments and theoretical analyses on a linear model. On CIFAR10, it boosts the state-of-the-art model from 66.16% to 72.66% against the four attacks of AutoAttack, including 71.76% to 83.30% against the Square attack. On ImageNet, the top-1 robust accuracy of FastAT is improved from 33.18% to 38.54% under the 100-step PGD attack.

Optical remote sensing imagery has been widely used in many fields due to its high resolution and stable geometric properties. However, remote sensing imagery is inevitably affected by climate, especially clouds. Removing the cloud in the high-resolution remote sensing satellite image is an indispensable pre-processing step before analyzing it. For the sake of large-scale training data, neural networks have been successful in many image processing tasks, but the use of neural networks to remove cloud in remote sensing imagery is still relatively small. We adopt generative adversarial network to solve this task and introduce the spatial attention mechanism into the remote sensing imagery cloud removal task, proposes a model named spatial attention generative adversarial network (SpA GAN), which imitates the human visual mechanism, and recognizes and focuses the cloud area with local-to-global spatial attention, thereby enhancing the information recovery of these areas and generating cloudless images with better quality...

Phase change material (PCM) features distinct optical or electronic properties between amorphous and crystalline states. Recently, it starts to play a key role in the emerging photonic applications like optoelectronic display, dynamic wavefront control, on-chip photonic memory and computation. However, current PCMs do not refract effectively at visible wavelengths and suffer from deformation and decomposition, limiting the repeatability and vast visible wavelength applications. Silicon as the fundamental material for electronics and photonics, has never been considered as phase change material, due to its ultrafast crystallization kinetics. Here we show the striking fact that nanoscale silicon domains can be reversibly crystallized and amorphized under nanosecond laser pulses. For a typical disk resonator, it also provides a 25% non-volatile modulation at nanosecond time scale. We further show proof-of-concept experiments that such attributes could enable ultra-high resolution dielectric color display and dynamic visible wavefront control.