no code implementations • 15 May 2024 • Li Ma, Yifan Zhao, Peixi Peng, Yonghong Tian
Different from these methods, we propose to decouple the intrinsic attributes into two complementary features for artifacts reduction, ie, the compression-insensitive features to regularize the high-level semantic representations during training and the compression-sensitive features to be aware of the compression degree.
no code implementations • 1 Apr 2024 • Liwen Zhu, Peixi Peng, Zongqing Lu, Yonghong Tian
Traffic signal control has a great impact on alleviating traffic congestion in modern cities.
no code implementations • 7 Mar 2024 • Ding Chen, Peixi Peng, Tiejun Huang, Yonghong Tian
As a general method for exploration in deep reinforcement learning (RL), NoisyNet can produce problem-specific exploration strategies.
no code implementations • 6 Mar 2024 • Guangyao Chen, Peixi Peng, Yangru Huang, Mengyue Geng, Yonghong Tian
One important desideratum of lifelong learning aims to discover novel classes from unlabelled data in a continuous manner.
no code implementations • 9 Jan 2024 • Ding Chen, Peixi Peng, Tiejun Huang, Yonghong Tian
Recently, the surrogate gradient method has been utilized for training multi-layer SNNs, which allows SNNs to achieve comparable performance with the corresponding deep networks in this task.
no code implementations • 14 Jul 2023 • Mingjian Ni, Guangyao Chen, Xiawu Zheng, Peixi Peng, Li Yuan, Yonghong Tian
Applying such theory, we propose a plug-and-play CKA-based Sparsity Regularization for sparse network training, dubbed CKA-SR, which utilizes CKA to reduce feature similarity between layers and increase network sparsity.
no code implementations • 8 Jun 2023 • Yunpeng Zhai, Peixi Peng, Mengxi Jia, Shiyong Li, Weiqiang Chen, Xuesong Gao, Yonghong Tian
Extensive experiments demonstrate that (1) CRS approximately measures the performance of models without labeled samples; (2) and PEG produces new state-of-the-art accuracy for person re-identification, indicating the great potential of population-based network cooperative training for unsupervised learning.
Knowledge Distillation Unsupervised Person Re-Identification
1 code implementation • 21 Apr 2023 • Li Ma, Peixi Peng, Guangyao Chen, Yifan Zhao, Siwei Dong, Yonghong Tian
The sensitivity of deep neural networks to compressed images hinders their usage in many real applications, which means classification networks may fail just after taking a screenshot and saving it as a compressed file.
1 code implementation • CVPR 2023 • Zeyin Song, Yifan Zhao, Yujun Shi, Peixi Peng, Li Yuan, Yonghong Tian
However, in this work, we find that the CE loss is not ideal for the base session training as it suffers poor class separation in terms of representations, which further degrades generalization to novel classes.
2 code implementations • 27 Jan 2023 • Guangyao Chen, Peixi Peng, Guoqi Li, Yonghong Tian
The accumulation in AAP could compensate for the information loss during the forward and backward of full spike propagation, and facilitate the training of the FSNN.
no code implementations • ICCV 2023 • Yangru Huang, Peixi Peng, Yifan Zhao, Yunpeng Zhai, Haoran Xu, Yonghong Tian
Efficient motion and appearance modeling are critical for vision-based Reinforcement Learning (RL).
no code implementations • ICCV 2023 • Yunpeng Zhai, Peixi Peng, Yifan Zhao, Yangru Huang, Yonghong Tian
Vision-based reinforcement learning (RL) depends on discriminative representation encoders to abstract the observation states.
no code implementations • 10 Mar 2022 • Lantian Xue, Yixiong Zou, Peixi Peng, Yonghong Tian, Tiejun Huang
To solve this problem, we propose the Annotation Efficient Person Re-Identification method to select image pairs from an alternative pair set according to the fallibility and diversity of pairs, and train the Re-ID model based on the annotation.
no code implementations • 21 Jan 2022 • Ding Chen, Peixi Peng, Tiejun Huang, Yonghong Tian
With the help of special neuromorphic hardware, spiking neural networks (SNNs) are expected to realize artificial intelligence (AI) with less energy consumption.
1 code implementation • ICCV 2021 • Guangyao Chen, Peixi Peng, Li Ma, Jia Li, Lin Du, Yonghong Tian
This observation leads to more explanations of the CNN's generalization behaviors in both robustness to common perturbations and out-of-distribution detection, and motivates a new perspective on data augmentation designed by re-combing the phase spectrum of the current image and the amplitude spectrum of the distracter image.
Ranked #7 on Out-of-Distribution Detection on CIFAR-10
1 code implementation • 1 Mar 2021 • Guangyao Chen, Peixi Peng, Xiangqian Wang, Yonghong Tian
Then, an adversarial margin constraint is proposed to reduce the open space risk by limiting the latent open space constructed by reciprocal points.
3 code implementations • 4 Jan 2021 • Liwen Zhu, Peixi Peng, Zongqing Lu, Xiangqian Wang, Yonghong Tian
To make the policy learned from a training scenario generalizable to new unseen scenarios, a novel Meta Variationally Intrinsic Motivated (MetaVIM) RL method is proposed to learn the decentralized policy for each intersection that considers neighbor information in a latent way.
1 code implementation • ECCV 2020 • Guangyao Chen, Limeng Qiao, Yemin Shi, Peixi Peng, Jia Li, Tiejun Huang, ShiLiang Pu, Yonghong Tian
In this process, one of the key challenges is to reduce the risk of generalizing the inherent characteristics of numerous unknown samples learned from a small amount of known data.
1 code implementation • Proceedings of the 28th ACM International Conference on Multimedia 2020 • Feifei Ding, Peixi Peng, Yangru Huang, Mengyue Geng, Yonghong Tian
The proposed LPD model is trained in an end-to-end manner and only utilizes the original and synthetic training data.
no code implementations • 25 May 2019 • Yangru Huang, Peixi Peng, Yi Jin, Junliang Xing, Congyan Lang, Songhe Feng
To reduce domain divergence caused by that the source and target datasets are collected from different environments, we force to project the DSH feature maps from different domains to a new nominal domain, and a novel domain similarity loss is proposed based on one-class classification.
no code implementations • 5 Dec 2018 • Peixi Peng, Junliang Xing
To learn the multi-agent cooperation effectively and tackle the sub-optimality of demonstration, a self-improving learning method is proposed: On the one hand, the centralized state-action values are initialized by the demonstration and updated by the learned decentralized policy to improve the sub-optimality.
no code implementations • ECCV 2018 • Mengdan Zhang, Qiang Wang, Junliang Xing, Jin Gao, Peixi Peng, Weiming Hu, Steve Maybank
Correlation filters based trackers rely on a periodic assumption of the search sample to efficiently distinguish the target from the background.
no code implementations • CVPR 2016 • Peixi Peng, Tao Xiang, Yao-Wei Wang, Massimiliano Pontil, Shaogang Gong, Tiejun Huang, Yonghong Tian
Most existing person re-identification (Re-ID) approaches follow a supervised learning framework, in which a large number of labelled matching pairs are required for training.