no code implementations • 15 Apr 2024 • Guangyan Li, Yongqiang Tang, Wensheng Zhang
With this regard, we design a mixed compression model, which organically combines Low-Rank matrix approximation And structured Pruning (LoRAP).
no code implementations • 3 May 2023 • Ding Li, Yongqiang Tang, Zhizhong Zhang, Wensheng Zhang
Besides, to further exploit the potential of positive pairs and increase the robustness of self-supervised representation learning, we propose a Positive Feature Transformation (PFT) strategy which adopts feature-level manipulation to increase the variance of positive pairs.
no code implementations • 17 Feb 2023 • Lei Tian, Yongqiang Tang, Liangchen Hu, Wensheng Zhang
In this paper, we propose a novel domain adaptation method, which infers target pseudo-labels through cross-domain label propagation, such that the underlying manifold structure of two domain data can be explored.
no code implementations • 9 Oct 2022 • Jianxin Lin, Yongqiang Tang, JunPing Wang, Wensheng Zhang
Finally, a Constrained Maximum Cross-domain Likelihood (CMCL) optimization problem is deduced, by solving which the joint distributions are naturally aligned.
no code implementations • 17 Sep 2022 • Jianxin Lin, Yongqiang Tang, JunPing Wang, Wensheng Zhang
Domain generalization (DG) aims to learn a model on several source domains, hoping that the model can generalize well to unseen target domains.
no code implementations • 31 Aug 2022 • Ding Li, Xuebing Yang, Yongqiang Tang, Chenyang Zhang, Wensheng Zhang
And the other introduces a new metric based on mutual information between adjacent action proposals and evaluates the informativeness of video samples, named Temporal Context Inconsistency (TCI).
no code implementations • 31 Aug 2022 • Ding Li, Yuan Xie, Wensheng Zhang, Yongqiang Tang, Zhizhong Zhang
However, the existing methods simply employed max/average pooling in this framework, which ignored the distinct contributions of different individuals to the group activity recognition.
no code implementations • 10 Jun 2022 • Rui Chen, Yongqiang Tang, Wensheng Zhang, Wenlong Feng
Multi-view clustering has attracted much attention thanks to the capacity of multi-source information integration.
no code implementations • 4 Nov 2021 • Ding Li, Rui Wu, Yongqiang Tang, Zhizhong Zhang, Wensheng Zhang
Specifically, we first construct a two-dimensional map for each temporal scale to capture the temporal dependencies between candidates.
1 code implementation • 20 Mar 2020 • Lei Tian, Yongqiang Tang, Liangchen Hu, Zhida Ren, Wensheng Zhang
Domain adaptation has been a fundamental technology for transferring knowledge from a source domain to a target domain.