Papers › CoLA: Weakly-Supervised Temporal Action Localization with Snippet Contrastive Learning

CoLA: Weakly-Supervised Temporal Action Localization with Snippet Contrastive Learning

30 Mar 2021CVPR 2021 1arXiv:2103.16392archive 2025-07-28

Can Zhang, Meng Cao, Dongming Yang, Jie Chen, Yuexian Zou

Weakly-supervised temporal action localization (WS-TAL) aims to localize actions in untrimmed videos with only video-level labels. Most existing models follow the "localization by classification" procedure: locate temporal regions contributing most to the video-level classification. Generally, they process each snippet (or frame) individually and thus overlook the fruitful temporal context relation. Here arises the single snippet cheating issue: "hard" snippets are too vague to be classified. In this paper, we argue that learning by comparing helps identify these hard snippets and we propose to utilize snippet Contrastive learning to Localize Actions, CoLA for short. Specifically, we propose a Snippet Contrast (SniCo) Loss to refine the hard snippet representation in feature space, which guides the network to perceive precise temporal boundaries and avoid the temporal interval interruption. Besides, since it is infeasible to access frame-level annotations, we introduce a Hard Snippet Mining algorithm to locate the potential hard snippets. Substantial analyses verify that this mining strategy efficaciously captures the hard snippets and SniCo Loss leads to more informative feature representation. Extensive experiments show that CoLA achieves state-of-the-art results on THUMOS'14 and ActivityNet v1.2 datasets. CoLA code is publicly available at https://github.com/zhang-can/CoLA.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

zhang-can/CoLA officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Action LocalizationContrastive LearningTemporal Action LocalizationWeakly Supervised Action LocalizationWeakly-supervised Temporal Action Localization

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Weakly Supervised Action Localization ActivityNet-1.2 CoLA Mean mAP 26.1 #7 of 19 Archive leaderboard report
Weakly Supervised Action Localization ActivityNet-1.2 CoLA mAP@0.5 42.7 #7 of 19 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 CoLA mAP@0.1:0.5 50.3 #16 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 CoLA mAP@0.1:0.7 40.9 #16 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 CoLA mAP@0.5 32.2 #16 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS14 CoLA avg-mAP (0.1-0.5) 50.3 #9 of 12 Archive leaderboard report
Weakly Supervised Action Localization THUMOS14 CoLA avg-mAP (0.1:0.7) 40.9 #9 of 12 Archive leaderboard report
Weakly Supervised Action Localization THUMOS14 CoLA avg-mAP (0.3-0.7) 32.1 #9 of 12 Archive leaderboard report
Weakly Supervised Action Localization THUMOS’14 CoLA mAP@0.5 32.2 #9 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

COLAContrastive Learning

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections