activity
20182023
most citedA Unified Model for Multi-class Anomaly Detection

87 citations · 468 across the 50 of their papers we have counts for

collaborators
Showing 2021Show all

9 papers · 1 filter

cs.CV2021★ 2 cited

3D-aware Image Synthesis via Learning Structural and Textural Representations

Yinghao Xu, Sida Peng, Ceyuan Yang +2

Making generative models 3D-aware bridges the 2D image space and the 3D physical world yet remains challenging. Recent attempts equip a Generative Adversarial Network (GAN) with a…

cs.CV2021★ 6 cited

Cross-Model Pseudo-Labeling for Semi-Supervised Action Recognition

Yinghao Xu, Fangyun Wei, Xiao Sun +5

Semi-supervised action recognition is a challenging but important task due to the high cost of data annotation. A common approach to this problem is to assign unlabeled data with p…

cs.CV2021★ 5 cited

Improving GAN Equilibrium by Raising Spatial Awareness

Jianyuan Wang, Ceyuan Yang, Yinghao Xu +3

The success of Generative Adversarial Networks (GANs) is largely built upon the adversarial training between a generator (G) and a discriminator (D). They are expected to reach a c…

cs.CV2021★ 8 cited

One-Shot Generative Domain Adaptation

Ceyuan Yang, Yujun Shen, Zhiyi Zhang +4

This work aims at transferring a Generative Adversarial Network (GAN) pre-trained on one image domain to a new domain referring to as few as just one target image. The main challen…

cs.CV2021

CompConv: A Compact Convolution Module for Efficient Feature Learning

Chen Zhang, Yinghao Xu, Yujun Shen

Convolutional Neural Networks (CNNs) have achieved remarkable success in various computer vision tasks but rely on tremendous computational cost. To solve this problem, existing ap…

cs.CV2021★ 31 cited

Data-Efficient Instance Generation from Instance Discrimination

Ceyuan Yang, Yujun Shen, Yinghao Xu +1

Generative Adversarial Networks (GANs) have significantly advanced image synthesis, however, the synthesis quality drops significantly given a limited amount of training data. To i…