no code implementations • 26 Jun 2022 • Xiang Gao, Yingjie Tian, Zhiquan Qi
We propose an end-to-end-trainable feature augmentation module built for image classification that extracts and exploits multi-view local features to boost model performance.
no code implementations • 27 Sep 2021 • Jie Yang, Ruijie Xu, Zhiquan Qi, Yong Shi
Visual anomaly detection is an important and challenging problem in the field of machine learning and computer vision.
2 code implementations • 13 Sep 2021 • Kai Li, Jie Yang, Siwei Ma, Bo wang, Shanshe Wang, Yingjie Tian, Zhiquan Qi
For the second issue, we reconsider how to improve detection efficiency with excellent performance, and then propose our lightweight encoder-decoder architecture termed CarNet.
no code implementations • 29 Jun 2021 • Kai Li, Bo wang, Yingjie Tian, Zhiquan Qi
Numerous detection problems in computer vision, including road crack detection, suffer from exceedingly foreground-background imbalance.
no code implementations • 22 May 2021 • Jiabin Liu, Bo wang, Xin Shen, Zhiquan Qi, Yingjie Tian
Learning from label proportions (LLP) aims at learning an instance-level classifier with label proportions in grouped training data.
1 code implementation • 13 Dec 2020 • Jie Yang, Yong Shi, Zhiquan Qi
Concretely, we develop a multi-scale regional feature generator that can generate multiple spatial context-aware representations from pre-trained deep convolutional networks for every subregion of an image.
Ranked #67 on Anomaly Detection on MVTec AD
1 code implementation • 11 Feb 2020 • Jie Yang, Zhiquan Qi, Yong Shi
This paper develops a multi-task learning framework that attempts to incorporate the image structure knowledge to assist image inpainting, which is not well explored in previous works.
no code implementations • ICLR 2020 • Yong Shi, Biao Li, Bo wang, Zhiquan Qi, Jiabin Liu, Fan Meng
Super Resolution (SR) is a fundamental and important low-level computer vision (CV) task.
1 code implementation • 24 Sep 2019 • Biao Li, Jiabin Liu, Bo Wang, Zhiquan Qi, Yong Shi
Deep learning (DL) architectures for superresolution (SR) normally contain tremendous parameters, which has been regarded as the crucial advantage for obtaining satisfying performance.
1 code implementation • NeurIPS 2019 • Jiabin Liu, Bo wang, Zhiquan Qi, Yingjie Tian, Yong Shi
In this paper, we leverage generative adversarial networks (GANs) to derive an effective algorithm LLP-GAN for learning from label proportions (LLP), where only the bag-level proportional information in labels is available.
no code implementations • IEEE Transactions on Intelligent Transportation Systems 2016 • Yong Shi, Limeng Cui, Zhiquan Qi, Fan Meng, and Zhensong Chen
Our contributions are shown as follows: 1) apply the integral channel features to redefine the tokens that constitute a crack and get better representation of the cracks with intensity inhomogeneity; 2) introduce random structured forests to generate a high- performance crack detector, which can identify arbitrarily com- plex cracks; and 3) propose a new crack descriptor to characterize cracks and discern them from noises effectively.