Thomas S. Huang
Professor
Video instance segmentation is a complex task in which we need to detect...
Cardiac motion estimation plays a key role in MRI cardiac feature tracki...
Inspired by the robustness and efficiency of sparse representation in sp...
Deep convolution-based single image super-resolution (SISR) networks emb...
Self-similarity refers to the image prior widely used in image restorati...
The first Agriculture-Vision Challenge aims to encourage research in
dev...
Learning segmentation from synthetic data and adapting to real data can
...
While deep neural networks have been shown to perform remarkably well in...
We consider the problem of unsupervised domain adaptation for semantic
s...
Most of the modern instance segmentation approaches fall into two catego...
The success of deep learning in visual recognition tasks has driven
adva...
While scale-invariant modeling has substantially boosted the performance...
In this work, we introduce Panoptic-DeepLab, a simple, strong, and fast
...
We present Any-Precision Deep Neural Networks (Any-Precision DNNs), whic...
We present Panoptic-DeepLab, a bottom-up and single-shot approach for
pa...
Recently image-to-image translation has attracted significant interests ...
The one-to-one mapping is necessary for many bidirectional image-to-imag...
In this paper, we are interested in bottom-up multi-person human pose
es...
The training method of repetitively feeding all samples into a pre-defin...
Image denoising and high-level vision tasks are usually handled independ...
We present a novel deep learning based image inpainting system to comple...
Many classic methods have shown non-local self-similarity in natural ima...
Despite the remarkable progress, weakly supervised segmentation approach...
Advances in image super-resolution (SR) have recently benefited signific...
Recent deep learning based approaches have shown promising results on im...
Visual recognition under adverse conditions is a very important and
chal...
In this paper, we propose gated recurrent feature pyramid for the proble...
Learning with recurrent neural networks (RNNs) on long sequences is a
no...
Emotion recognition from facial expressions is tremendously useful,
espe...
Similarity-based clustering and semi-supervised learning methods separat...
Conventionally, image denoising and high-level vision tasks are handled
...
Convolutional autoregressive models have recently demonstrated
state-of-...
Learning from weakly-supervised data is one of the main challenges in ma...
This paper emphasizes the significance to jointly exploit the problem
st...
With the agreement of my coauthors, I Zhangyang Wang would like to withd...
We investigate the ℓ_∞-constrained representation which
demonstrates rob...
Video object detection is challenging because objects that are easily
de...
We consider the task of dimensional emotion recognition on video data us...
Image aesthetics assessment has been challenging due to its subjective
n...
Visual recognition research often assumes a sufficient resolution of the...
In this paper, we design a Deep Dual-Domain (D^3) based fast
restoration...
Despite being the appearance-based classifier of choice in recent years,...
Despite its nonconvex nature, ℓ_0 sparse approximation is desirable in
m...
While sparse coding-based clustering methods have shown to be successful...
As font is one of the core design concepts, automatic font identificatio...
Deep learning has been successfully applied to image super resolution (S...
We address a challenging fine-grain classification problem: recognizing ...
We study the complementary behaviors of external and internal examples i...
Single image super-resolution (SR) aims to estimate a high-resolution (H...
Convolutional neural networks perform well on object recognition because...