Mengyuan Chen

Mengyuan Chen

Auto-generated profile, based on publicly available information

Mengyuan Chen

State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS), Institute of Automation, Chinese Academy of Sciences, Beijing, China

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
4 September 2023 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
JGJunyu Gao
Junyu Gao
MCMengyuan Chen
Mengyuan Chen
CXChangsheng Xu
Changsheng Xu
With the explosive growth of videos, weakly-supervised temporal action localization (WS-TAL) task has become a promising research direction in pattern analysis and machine learning. WS-TAL aims to detect and localize action instances with only video-level labels during training. Modern approaches have achieved impressive progress via powerful deep neural networks. However, robust and reliable WS-TAL remains challenging and underexplored due to considerable uncertainty caused by weak supervision, noisy evaluation environment, and unknown categories in the open world. To this end, we propose a new paradigm, named vectorized evidential learning (VEL), to explore local-to-global evidence collection for facilitating model performance. Specifically, a series of learnable meta-action units (MAUs) are automatically constructed, which serve as fundamental elements constituting diverse action categories. Since the same meta-action unit can manifest as distinct action components within different action categories, we leverage MAUs and category representations to dynamically and adaptively learn action components and action-component relations. After performing uncertainty estimation at both category-level and unit-level, the local evidence from action components is accumulated and optimized under the Subject Logic theory. Extensive experiments on the regular, noisy, and open-set settings of three popular benchmarks show that VEL consistently obtains more robust and reliable action localization performance than state-of-the-arts.

28 October 2022
Verified
Book Chapter
Closed access
PDF file
MCMengyuan Chen
Mengyuan Chen
JGJunyu Gao
Junyu Gao
SYShicai Yang
Shicai Yang
CXChangsheng Xu
Changsheng Xu

26 February 2025 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
JGJunyu Gao
Junyu Gao
MCMengyuan Chen
Mengyuan Chen
CXChangsheng Xu
Changsheng Xu
With only video-level event labels, this paper targets at the task of weakly-supervised audio-visual event perception (WS-AVEP), which aims to temporally localize and categorize events that belong to each modality. Despite the recent progress, most existing approaches either ignore the unsynchronized property of audio-visual tracks or discount the complementary modality for explicit enhancement. We argue that, a modality should provide ample presence evidence for an event, while the complementary modality offers absence evidence as a reference. However, to learn reliable evidence, we face challenging uncertainties caused by weak supervision and the complicated audio-visual data itself. To this end, we propose to collect Probabilistic Presence-Absence Evidence (PPAE) in a unified framework. Specifically, by leveraging uni-modal and cross-modal representations, a probabilistic presence-absence evidence collector (PAEC) is designed. To learn the evidence in a reliable range, we propose a joint-modal mutual learning (JML) process, which calibrates the evidence of diverse audible, visible, and audi-visible events adaptively and dynamically. Extensive experiments show that our method surpasses state-of-the-arts (e.g., absolute gains of $3.1\%$ and $4.2\%$ in terms of event-level audio and visual metrics on the LLP dataset).

25 August 2023 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
MCMengyuan Chen
Mengyuan Chen
JGJunyu Gao
Junyu Gao
CXChangsheng Xu
Changsheng Xu
Weakly-supervised temporal action localization (WTAL) aims to localize the action instances and recognize their categories with only video-level labels. Despite great progress, existing methods suffer from severe action-background ambiguity, which mainly arises from background noise and neglect of non-salient action snippets. To address this issue, we propose a generalized evidential deep learning (EDL) framework for WTAL, called Uncertainty-aware Dual-Evidential Learning (UDEL), which extends the traditional paradigm of EDL to adapt to the weakly-supervised multi-label classification goal with the guidance of epistemic and aleatoric uncertainties of which the former comes from models lacking knowledge, while the latter comes from the inherent properties of samples themselves. Specifically, targeting excluding the undesirable background snippets, we fuse the video-level epistemic and aleatoric uncertainties to measure the interference of background noise to video-level prediction. Then, the snippet-level aleatoric uncertainty is further deduced for progressive mutual learning, which gradually focuses on the entire action instances in an “easy-to-hard” manner and encourages the snippet-level epistemic uncertainty to be complementary with the foreground attention scores. Extensive experiments show that UDEL achieves state-of-the-art performance on four public benchmarks. Our code is available in github/mengyuanchen2021/UDEL .

6 June 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
YHYufan Hu
Yufan Hu
JFJie Fu
Jie Fu
MCMengyuan Chen
Mengyuan Chen
JGJunyu Gao
Junyu Gao
JDJianfeng Dong
Jianfeng Dong
HLHongmin Liu
Hongmin Liu
Weakly-supervised temporal action localization (WTAL) aims to localize and classify action instances in untrimmed videos with only video-level labels available. Despite the remarkable success of existing methods, whose generated proposals are commonly far more than the ground-truth action instances, it still makes sense to improve the ranking accuracy of the generated proposals since users in real-world scenarios usually prioritize the action proposals with the highest confidence scores. The inaccuracy of the proposal ranking mainly comes from two aspects: For one thing, the traditional proposal generation manner entirely relies on snippet-level perception, resulting in a significant yet unnoticed gap with the target of proposal-level localization. For another, existing methods commonly employ a hand-crafted proposal generation manner, a post-process that does not participate in model optimization. To address the above issues, we propose an end-to-end trained two-stage method, termed as Learning Proposal-aware Re-ranking (LPR) for WTAL. In the first stage, we design a proposal-aware feature learning module to inject the proposal-aware contextual information into each snippet, and then the enhanced features are utilized for predicting initial proposals. Furthermore, to perform effective and efficient proposal re-ranking, in the second stage, we contrast the proposals attached with high confidence scores with our constructed multi-scale foreground/background prototypes for further optimization. Evaluated by both the vanilla and Top- k mAP metrics, results of extensive experiments on two popular benchmarks demonstrate the effectiveness of our proposed method.

6 June 2021
Verified
Conference Paper
Closed access
XCXu Chen
Xu Chen
LDLun Du
Lun Du
MCMengyuan Chen
Mengyuan Chen
YWYun Wang
Yun Wang
QLQingqing Long
Qingqing Long
KXKunqing Xie
Kunqing Xie
Hierarchy preserving network embedding is a method that project nodes into feature space by preserving the hierarchy property of networks. Recently, researches on network representation have considerably profited from taking hierarchy into consideration. Among these works, SpaceNE 1 [1] stands out by preserving hierarchy with the help of subspace constraints on the hierarchy subspace system. However, like all other hierarchy preserving network embedding methods, SpaceNE is time-consuming and cannot generalize to new nodes. In this paper, we propose an inductive method, FastHGE, to learn node representations more efficiently and generalize to new nodes more easily. Empirically, the experiment of node classification demonstrates that the convergence speed of FastHGE is increased by 30 times in the case of the same accuracy with SpaceNE.