arXiv subjects

Bin Fu

Publications and source records attributed to Bin Fu.

At least 91 records · Page 5Linked to original sources

arXiv2023

Enhancing Quality of Pose-varied Face Restoration with Local Weak Feature Sensing and GAN Prior

Facial semantic guidance (including facial landmarks, facial heatmaps, and facial parsing maps) and facial generative adversarial networks (GAN) prior have been widely used in blind face restoration (BFR) in recent years. Although existing BFR methods have achieved good performance in ordinary cases, these solutions have limited resilience when applied to face images with serious degradation and pose-varied (e.g., looking right, looking left, laughing, etc.) in real-world scenarios. In this work, we propose a well-designed blind face restoration network with generative facial prior. The proposed network is mainly comprised of an asymmetric codec and a StyleGAN2 prior network. In the asymmetric codec, we adopt a mixed multi-path residual block (MMRB) to gradually extract weak texture features of input images, which can better preserve the original facial features and avoid excessive fantasy. The MMRB can also be plug-and-play in other networks. Furthermore, thanks to the affluent and diverse facial priors of the StyleGAN2 model, we adopt it as the primary generator network in our proposed method and specially design a novel self-supervised training strategy to fit the distribution closer to the target and flexibly restore natural and realistic facial details. Extensive experiments on synthetic and real-world datasets demonstrate that our model performs superior to the prior art for face restoration and face super-resolution tasks.

cs.CV↗

arXiv2023

StyleAvatar3D: Leveraging Image-Text Diffusion Models for High-Fidelity 3D Avatar Generation

The recent advancements in image-text diffusion models have stimulated research interest in large-scale 3D generative models. Nevertheless, the limited availability of diverse 3D resources presents significant challenges to learning. In this paper, we present a novel method for generating high-quality, stylized 3D avatars that utilizes pre-trained image-text diffusion models for data generation and a Generative Adversarial Network (GAN)-based 3D generation network for training. Our method leverages the comprehensive priors of appearance and geometry offered by image-text diffusion models to generate multi-view images of avatars in various styles. During data generation, we employ poses extracted from existing 3D models to guide the generation of multi-view images. To address the misalignment between poses and images in data, we investigate view-specific prompts and develop a coarse-to-fine discriminator for GAN training. We also delve into attribute-related prompts to increase the diversity of the generated avatars. Additionally, we develop a latent diffusion model within the style space of StyleGAN to enable the generation of avatars based on image inputs. Our approach demonstrates superior performance over current state-of-the-art methods in terms of visual quality and diversity of the produced avatars.

cs.CV↗

arXiv2023

Executing your Commands via Motion Diffusion in Latent Space

We study a challenging task, conditional human motion generation, which produces plausible human motion sequences according to various conditional inputs, such as action classes or textual descriptors. Since human motions are highly diverse and have a property of quite different distribution from conditional modalities, such as textual descriptors in natural languages, it is hard to learn a probabilistic mapping from the desired conditional modality to the human motion sequences. Besides, the raw motion data from the motion capture system might be redundant in sequences and contain noises; directly modeling the joint distribution over the raw motion sequences and conditional modalities would need a heavy computational overhead and might result in artifacts introduced by the captured noises. To learn a better representation of the various human motion sequences, we first design a powerful Variational AutoEncoder (VAE) and arrive at a representative and low-dimensional latent code for a human motion sequence. Then, instead of using a diffusion model to establish the connections between the raw motion sequences and the conditional inputs, we perform a diffusion process on the motion latent space. Our proposed Motion Latent-based Diffusion model (MLD) could produce vivid motion sequences conforming to the given conditional inputs and substantially reduce the computational overhead in both the training and inference stages. Extensive experiments on various human motion generation tasks demonstrate that our MLD achieves significant improvements over the state-of-the-art methods among extensive human motion generation tasks, with two orders of magnitude faster than previous diffusion models on raw motion sequences.

cs.CV↗

arXiv2023

Sublinear Approximation Schemes for Scheduling Precedence Graphs of Bounded Depth

We study the classical scheduling problem on parallel machines %with precedence constraints where the precedence graph has the bounded depth $h$. Our goal is to minimize the maximum completion time. We focus on developing approximation algorithms that use only sublinear space or sublinear time. We develop the first one-pass streaming approximation schemes using sublinear space when all jobs' processing times differ no more than a constant factor $c$ and the number of machines $m$ is at most $\tfrac {2n ε}{3 h c }$. This is so far the best approximation we can have in terms of $m$, since no polynomial time approximation better than $\tfrac{4}{3}$ exists when $m = \tfrac{n}{3}$ unless P=NP. %the problem cannot be approximated within a factor of $\tfrac{4}{3}$ when $m = \tfrac{n}{3}$ even if all jobs have equal processing time. The algorithms are then extended to the more general problem where the largest $αn$ jobs have no more than $c$ factor difference. % for some constant $0 < α\le 1$. We also develop the first sublinear time algorithms for both problems. For the more general problem, when $ m \le \tfrac { αn ε}{20 c^2 \cdot h } $, our algorithm is a randomized $(1+ε)$-approximation scheme that runs in sublinear time. This work not only provides an algorithmic solution to the studied problem under big data % and cloud computing environment, but also gives a methodological framework for designing sublinear approximation algorithms for other scheduling problems.

cs.DS↗

arXiv2022

Efficient Single-Image Depth Estimation on Mobile Devices, Mobile AI & AIM 2022 Challenge: Report

Various depth estimation models are now widely used on many mobile and IoT devices for image segmentation, bokeh effect rendering, object tracking and many other mobile tasks. Thus, it is very crucial to have efficient and accurate depth estimation models that can run fast on low-power mobile chipsets. In this Mobile AI challenge, the target was to develop deep learning-based single image depth estimation solutions that can show a real-time performance on IoT platforms and smartphones. For this, the participants used a large-scale RGB-to-depth dataset that was collected with the ZED stereo camera capable to generated depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the Raspberry Pi 4 platform, where the developed solutions were able to generate VGA resolution depth maps at up to 27 FPS while achieving high fidelity results. All models developed in the challenge are also compatible with any Android or Linux-based mobile devices, their detailed description is provided in this paper.

cs.CV↗

arXiv2022

Learning Variational Motion Prior for Video-based Motion Capture

Motion capture from a monocular video is fundamental and crucial for us humans to naturally experience and interact with each other in Virtual Reality (VR) and Augmented Reality (AR). However, existing methods still struggle with challenging cases involving self-occlusion and complex poses due to the lack of effective motion prior modeling. In this paper, we present a novel variational motion prior (VMP) learning approach for video-based motion capture to resolve the above issue. Instead of directly building the correspondence between the video and motion domain, We propose to learn a generic latent space for capturing the prior distribution of all natural motions, which serve as the basis for subsequent video-based motion capture tasks. To improve the generalization capacity of prior space, we propose a transformer-based variational autoencoder pretrained over marker-based 3D mocap data, with a novel style-mapping block to boost the generation quality. Afterward, a separate video encoder is attached to the pretrained motion generator for end-to-end fine-tuning over task-specific video datasets. Compared to existing motion prior models, our VMP model serves as a motion rectifier that can effectively reduce temporal jittering and failure modes in frame-wise pose estimation, leading to temporally stable and visually realistic motion capture results. Furthermore, our VMP-based framework models motion at sequence level and can directly generate motion clips in the forward pass, achieving real-time motion capture during inference. Extensive experiments over both public datasets and in-the-wild videos have demonstrated the efficacy and generalization capability of our framework.

cs.CV↗

arXiv2022

Hierarchical Normalization for Robust Monocular Depth Estimation

In this paper, we address monocular depth estimation with deep neural networks. To enable training of deep monocular estimation models with various sources of datasets, state-of-the-art methods adopt image-level normalization strategies to generate affine-invariant depth representations. However, learning with image-level normalization mainly emphasizes the relations of pixel representations with the global statistic in the images, such as the structure of the scene, while the fine-grained depth difference may be overlooked. In this paper, we propose a novel multi-scale depth normalization method that hierarchically normalizes the depth representations based on spatial information and depth distributions. Compared with previous normalization strategies applied only at the holistic image level, the proposed hierarchical normalization can effectively preserve the fine-grained details and improve accuracy. We present two strategies that define the hierarchical normalization contexts in the depth domain and the spatial domain, respectively. Our extensive experiments show that the proposed normalization strategy remarkably outperforms previous normalization methods, and we set new state-of-the-art on five zero-shot transfer benchmark datasets.

cs.CV↗

arXiv2022

GenText: Unsupervised Artistic Text Generation via Decoupled Font and Texture Manipulation

Automatic artistic text generation is an emerging topic which receives increasing attention due to its wide applications. The artistic text can be divided into three components, content, font, and texture, respectively. Existing artistic text generation models usually focus on manipulating one aspect of the above components, which is a sub-optimal solution for controllable general artistic text generation. To remedy this issue, we propose a novel approach, namely GenText, to achieve general artistic text style transfer by separably migrating the font and texture styles from the different source images to the target images in an unsupervised manner. Specifically, our current work incorporates three different stages, stylization, destylization, and font transfer, respectively, into a unified platform with a single powerful encoder network and two separate style generator networks, one for font transfer, the other for stylization and destylization. The destylization stage first extracts the font style of the font reference image, then the font transfer stage generates the target content with the desired font style. Finally, the stylization stage renders the resulted font image with respect to the texture style in the reference image. Moreover, considering the difficult data acquisition of paired artistic text images, our model is designed under the unsupervised setting, where all stages can be effectively optimized from unpaired data. Qualitative and quantitative results are performed on artistic text benchmarks, which demonstrate the superior performance of our proposed model. The code with models will become publicly available in the future.

cs.CV↗

arXiv2022

Streaming Algorithms for Multitasking Scheduling with Shared Processing

In this paper, we design the first streaming algorithms for the problem of multitasking scheduling on parallel machines with shared processing. In one pass, our streaming approximation schemes can provide an approximate value of the optimal makespan. If the jobs can be read in two passes, the algorithm can find the schedule with the approximate value. This work not only provides an algorithmic big data solution for the studied problem, but also gives an insight into the design of streaming algorithms for other problems in the area of scheduling.

cs.DS↗

arXiv2022

Streaming Approximation Scheme for Minimizing Total Completion Time on Parallel Machines Subject to Varying Processing Capacity

We study the problem of minimizing total completion time on parallel machines subject to varying processing capacity. In this paper, we develop an approximation scheme for the problem under the data stream model where the input data is massive and cannot fit into memory and thus can only be scanned for a few passes. Our algorithm can compute the approximate value of the optimal total completion time in one pass and output the schedule with the approximate value in two passes.

cs.DS↗

arXiv2022

Multitasking Scheduling with Shared Processing

Recently, the problem of multitasking scheduling has attracted a lot of attention in the service industries where workers frequently perform multiple tasks by switching from one task to another. Hall, Leung and Li (Discrete Applied Mathematics 2016) proposed a shared processing multitasking scheduling model which allows a team to continue to work on the primary tasks while processing the routinely scheduled activities as they occur. The processing sharing is achieved by allocating a fraction of the processing capacity to routine jobs and the remaining fraction, which we denote as sharing ratio, to the primary jobs. In this paper, we generalize this model to parallel machines and allow the fraction of the processing capacity assigned to routine jobs to vary from one to another. The objectives are minimizing makespan and minimizing the total completion time. We show that for both objectives, there is no polynomial time approximation algorithm unless P=NP if the sharing ratios are arbitrary for all machines. Then we consider the problems where the sharing ratios on some machines have a constant lower bound. For each objective, we analyze the performance of the classical scheduling algorithms and their variations and then develop a polynomial time approximation scheme when the number of machines is a constant.

cs.DS↗

arXiv2021

Fine-grained Identity Preserving Landmark Synthesis for Face Reenactment

Recent face reenactment works are limited by the coarse reference landmarks, leading to unsatisfactory identity preserving performance due to the distribution gap between the manipulated landmarks and those sampled from a real person. To address this issue, we propose a fine-grained identity-preserving landmark-guided face reenactment approach. The proposed method has two novelties. First, a landmark synthesis network which is designed to generate fine-grained landmark faces with more details. The network refines the manipulated landmarks and generates a smooth and gradually changing face landmark sequence with good identity preserving ability. Second, several novel loss functions including synthesized face identity preserving loss, foreground/background mask loss as well as boundary loss are designed, which aims at synthesizing clear and sharp high-quality faces. Experiments are conducted on our self-collected BeautySelfie and the public VoxCeleb1 datasets. The presented qualitative and quantitative results show that our method can reenact fine-grained higher quality faces with good ID-preserved appearance details, fewer artifacts and clearer boundaries than state-of-the-art works. Code will be released for reproduction.

cs.CV↗

arXiv2021

Shuffle Transformer with Feature Alignment for Video Face Parsing

This is a short technical report introducing the solution of the Team TCParser for Short-video Face Parsing Track of The 3rd Person in Context (PIC) Workshop and Challenge at CVPR 2021. In this paper, we introduce a strong backbone which is cross-window based Shuffle Transformer for presenting accurate face parsing representation. To further obtain the finer segmentation results, especially on the edges, we introduce a Feature Alignment Aggregation (FAA) module. It can effectively relieve the feature misalignment issue caused by multi-resolution feature aggregation. Benefiting from the stronger backbone and better feature aggregation, the proposed method achieves 86.9519% score in the Short-video Face Parsing track of the 3rd Person in Context (PIC) Workshop and Challenge, ranked the first place.

cs.CV↗

arXiv2021

Shuffle Transformer: Rethinking Spatial Shuffle for Vision Transformer

Very recently, Window-based Transformers, which computed self-attention within non-overlapping local windows, demonstrated promising results on image classification, semantic segmentation, and object detection. However, less study has been devoted to the cross-window connection which is the key element to improve the representation ability. In this work, we revisit the spatial shuffle as an efficient way to build connections among windows. As a result, we propose a new vision transformer, named Shuffle Transformer, which is highly efficient and easy to implement by modifying two lines of code. Furthermore, the depth-wise convolution is introduced to complement the spatial shuffle for enhancing neighbor-window connections. The proposed architectures achieve excellent performance on a wide range of visual tasks including image-level classification, object detection, and semantic segmentation. Code will be released for reproduction.

cs.CV↗

arXiv2021

Fast and Accurate Single-Image Depth Estimation on Mobile Devices, Mobile AI 2021 Challenge: Report

Depth estimation is an important computer vision problem with many practical applications to mobile devices. While many solutions have been proposed for this task, they are usually very computationally expensive and thus are not applicable for on-device inference. To address this problem, we introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based depth estimation solutions that can demonstrate a nearly real-time performance on smartphones and IoT platforms. For this, the participants were provided with a new large-scale dataset containing RGB-depth image pairs obtained with a dedicated stereo ZED camera producing high-resolution depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the popular Raspberry Pi 4 platform with a mobile ARM-based Broadcom chipset. The proposed solutions can generate VGA resolution depth maps at up to 10 FPS on the Raspberry Pi 4 while achieving high fidelity results, and are compatible with any Android or Linux-based mobile devices. A detailed description of all models developed in the challenge is provided in this paper.

eess.IV↗

arXiv2021

Software-Defined Edge Computing: A New Architecture Paradigm to Support IoT Data Analysis

The rapid deployment of Internet of Things (IoT) applications leads to massive data that need to be processed. These IoT applications have specific communication requirements on latency and bandwidth, and present new features on their generated data such as time-dependency. Therefore, it is desirable to reshape the current IoT architectures by exploring their inherent nature of communication and computing to support smart IoT data process and analysis. We introduce in this paper features of IoT data, trends of IoT network architectures, some problems in IoT data analysis, and their solutions. Specifically, we view that software-defined edge computing is a promising architecture to support the unique needs of IoT data analysis. We further present an experiment on data anomaly detection in this architecture, and the comparison between two architectures for ECG diagnosis. Results show that our method is effective and feasible.

cs.NI↗

arXiv2020

A Survey on Complex Question Answering over Knowledge Base: Recent Advances and Challenges

Question Answering (QA) over Knowledge Base (KB) aims to automatically answer natural language questions via well-structured relation information between entities stored in knowledge bases. In order to make KBQA more applicable in actual scenarios, researchers have shifted their attention from simple questions to complex questions, which require more KB triples and constraint inference. In this paper, we introduce the recent advances in complex QA. Besides traditional methods relying on templates and rules, the research is categorized into a taxonomy that contains two main branches, namely Information Retrieval-based and Neural Semantic Parsing-based. After describing the methods of these branches, we analyze directions for future research and introduce the models proposed by the Alime team.

cs.CL↗

arXiv2020

Hardness of Sparse Sets and Minimal Circuit Size Problem

We develop a polynomial method on finite fields to amplify the hardness of spare sets in nondeterministic time complexity classes on a randomized streaming model. One of our results shows that if there exists a $2^{n^{o(1)}}$-sparse set in $NTIME(2^{n^{o(1)}})$ that does not have any randomized streaming algorithm with $n^{o(1)}$ updating time, and $n^{o(1)}$ space, then $NEXP\not=BPP$, where a $f(n)$-sparse set is a language that has at most $f(n)$ strings of length $n$. We also show that if MCSP is $ZPP$-hard under polynomial time truth-table reductions, then $EXP\not=ZPP$.

cs.CC↗