Image restoration (IR) has been an indispensable and challenging task in...
Many common types of data can be represented as functions that map
coord...
In theory, vector quantization (VQ) is always better than scalar quantiz...
Despite a short history, neural image codecs have been shown to surpass
...
Image compression techniques typically focus on compressing rectangular
...
Image coding for machines (ICM) aims to compress images to support downs...
Modern image inpainting systems, despite the significant progress, often...
In recent years, we have witnessed the great advancement of Deep neural
...
Representing a signal as a continuous function parameterized by neural
n...
Vision transformer has demonstrated great potential in abundant vision t...
Large-scale vision-language models (VLMs) pre-trained on billion-level d...
Logical frameworks provide natural and direct ways of specifying and
rea...
Traditional representations for light fields can be separated into two t...
As a commonly-used image compression format, JPEG has been broadly appli...
Compressed Image Super-resolution has achieved great attention in recent...
Deep neural networks (DNNs) have shown great potential in non-reference ...
Existing learning-based methods for blind image quality assessment (BIQA...
Coronary CT Angiography (CCTA) is susceptible to various distortions (e....
Image Coding for Machines (ICM) aims to compress images for AI tasks ana...
Image compression has raised widespread interest recently due to its
sig...
Improving the generalization capability of Deep Neural Networks (DNNs) i...
This paper presents ActiveMLP, a general MLP-like backbone for computer
...
For deep reinforcement learning (RL) from pixels, learning effective sta...
Traditional media coding schemes typically encode image/video into a
sem...
In this paper, we present the first neural video codec that can compete ...
Data augmentation (DA) has been widely investigated to facilitate model
...
Confounders in deep learning are in general detrimental to model's
gener...
Collecting large clean-distorted training image pairs in real world is
n...
Unsupervised Person Re-identification (U-ReID) with pseudo labeling rece...
Despite the recent progress in light field super-resolution (LFSR) achie...
Learned video compression methods have demonstrated great promise in cat...
Unsupervised domain adaptive classification intends to improve
theclassi...
Learning good feature representations is important for deep reinforcemen...
Task-driven semantic video/image coding has drawn considerable attention...
Single image super-resolution (SISR) algorithms reconstruct high-resolut...
Recently, adaptive graph convolutional network based traffic prediction
...
Cloth-Changing person re-identification (CC-ReID) aims at matching the s...
Unsupervised domain adaptive (UDA) person re-identification (ReID) aims ...
For unsupervised domain adaptation (UDA), to alleviate the effect of dom...
Many unsupervised domain adaptation (UDA) methods exploit domain adversa...
Data augmentation (DA) plays a critical role in training deep neural net...
Video-based person re-identification (re-ID) aims at matching the same p...
Image-to-image translation (I2I) aims to transfer images from a source d...
For many practical computer vision applications, the learned models usua...
Learned image compression based on neural networks have made huge progre...
Traditional single image super-resolution (SISR) methods that focus on
s...
Image quality assessment (IQA) aims to estimate human perception based i...
Over the past several years, we have witnessed the impressive progress o...
Multi-choice Machine Reading Comprehension (MMRC) aims to select the cor...
Distributed training techniques have been widely deployed in large-scale...