Zhiheng Li

Dr. Zhiheng Li

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 May 2024 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
ZLZhiheng Li
Zhiheng Li
YZYujie Zhong
Yujie Zhong
RSRan Song
Ran Song
TLTianjiao Li
Tianjiao Li
LMLin Ma
Lin Ma
Wei ZhangWei Zhang
Wei Zhang
Pre-trained visual-language (ViL) models have demonstrated good zero-shot capability in video understanding tasks, where they were usually adapted through fine-tuning or temporal modeling. However, in the task of open-vocabulary temporal action localization (OV-TAL), such adaption reduces the robustness of ViL models against different data distributions, leading to a misalignment between visual representations and text descriptions of unseen action categories. As a result, existing methods often strike a trade-off between action detection and classification. Aiming at this issue, this paper proposes DeTAL, a simple but effective two-stage approach for OV-TAL. DeTAL decouples action detection from action classification to avoid the compromise between them, and the state-of-the-art methods for close-set action localization can be handily adapted to OV-TAL, which significantly improves the performance. Meanwhile, DeTAL can easily tackle the scenario where action category annotations are unavailable in the training dataset. In the experiments, we propose a new cross-dataset setting to evaluate the zero-shot capability of different methods. And the results demonstrate that DeTAL outperforms the state-of-the-art methods for OV-TAL on both THUMOS14 and ActivityNet1.3.

7 December 2022 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
ZLZhiheng Li
Zhiheng Li
XGXuyuan Gong
Xuyuan Gong
RSRan Song
Ran Song
PDPeng Duan
Peng Duan
JLJared Liu
Jared Liu
Wei ZhangWei Zhang
Wei Zhang
This paper focuses on skeleton-based few-shot action recognition. Since skeleton is essentially a sparse representation of human action, the feature maps extracted from it, through a standard encoder network in the few-shot condition, may not be sufficiently discriminative for some action sequences that look partially similar to each other. To address this issue, we propose a self and mutual adaptive matching (SMAM) module to convert such feature maps into more discriminative feature vectors. Our method, named as SMAM-Net, first leverages both the temporal information associated with each individual skeleton joint and the spatial relationship among them for feature extraction. Then, the SMAM module adaptively measures the similarity between labeled and query samples and further carries out feature matching within the query set to distinguish similar skeletons of various action categories. Experimental results show that the SMAM-Net outperforms other baselines on the large-scale NTU RGB+D 120 dataset in the tasks of one-shot and five-shot action recognition. We also report our results on smaller datasets including NTU RGB+D 60, SYSU and PKU-MMD to demonstrate that our method is reliable and generalises well on different datasets. Codes and the pretrained SMAM-Net will be made publicly available.

15 August 2021 in arXiv
Verified
Preprint
Open access
PDF file
TLTianjiao Li
Tianjiao Li
JLJun Liu
Jun Liu
Wei ZhangWei Zhang
Wei Zhang
YNYun Ni
Yun Ni
WWWenqian Wang
Wenqian Wang
ZLZhiheng Li
Zhiheng Li
Human behavior understanding with unmanned aerial vehicles (UAVs) is of great significance for a wide range of applications, which simultaneously brings an urgent demand of large, challenging, and comprehensive benchmarks for the development and evaluation of UAV-based models. However, existing benchmarks have limitations in terms of the amount of captured data, types of data modalities, categories of provided tasks, and diversities of subjects and environments. Here we propose a new benchmark - UAVHuman - for human behavior understanding with UAVs, which contains 67,428 multi-modal video sequences and 119 subjects for action recognition, 22,476 frames for pose estimation, 41,290 frames and 1,144 identities for person re-identification, and 22,263 frames for attribute recognition. Our dataset was collected by a flying UAV in multiple urban and rural districts in both daytime and nighttime over three months, hence covering extensive diversities w.r.t subjects, backgrounds, illuminations, weathers, occlusions, camera motions, and UAV flying attitudes. Such a comprehensive and challenging benchmark shall be able to promote the research of UAV-based human behavior understanding, including action recognition, pose estimation, re-identification, and attribute recognition. Furthermore, we propose a fisheye-based action recognition method that mitigates the distortions in fisheye videos via learning unbounded transformations guided by flat RGB videos. Experiments show the efficacy of our method on the UAV-Human dataset. The project page: https://github.com/SUTDCV/UAV-Human

9 November 2025
Verified
Book Chapter
Closed access
PDF file
YCYanyun Chen
Yanyun Chen
JSJiapeng Sheng
Jiapeng Sheng
XFXing Fang
Xing Fang
WTWenhao Tan
Wenhao Tan
ZLZhiheng Li
Zhiheng Li
Wei ZhangWei Zhang
Wei Zhang

23 March 2026 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
WTWenhao Tan
Wenhao Tan
XFXing Fang
Xing Fang
YCYanyun Chen
Yanyun Chen
ZLZhiheng Li
Zhiheng Li
RSRan Song
Ran Song
Wei ZhangWei Zhang
Wei Zhang
Bipedal locomotion is challenging for many learning-based algorithms. The sim-to-real gap, the changes in task loads, and the changes in joint damping caused by long-term use may all degrade the performance of these algorithms. Aiming at such issues, we propose a memory-informed meta-reinforcement learning (MI-MRL) method for bipedal locomotion. MI-MRL demonstrates robustness to various task-related parameter changes and offers a practical solution to the difficulties that MRL faces in real-world deployment. It replaces the standard feedback rewards commonly used in MRL and restructures the input to the context encoder, thus broadening the applicability of MRL-based methods to real-world scenarios. MI-MRL also introduces a state estimator and a noise filter to accelerate training and enhance algorithm performance. According to the experimental results, MI-MRL completes information collection and context adaptation within 0.32 s, and achieves an external load tolerance 20 times higher than that of the latest MRL-based methods. The results also demonstrate that MI-MRL is robust to various task-related parameter changes and offers a practical solution to the difficulties that MRL faces in real-world deployment.

1 January 2026 in IEEE Transactions on Automation Science and Engineering
Verified
Journal Article
Closed access
XFXing Fang
Xing Fang
ZLZhiheng Li
Zhiheng Li
YCYanyun Chen
Yanyun Chen
WTWenhao Tan
Wenhao Tan
MZMingxing Zhang
Mingxing Zhang
Wei ZhangWei Zhang
Wei Zhang