Papers › D2-Net: Weakly-Supervised Action Localization via Discriminative Embeddings and...

D2-Net: Weakly-Supervised Action Localization via Discriminative Embeddings and Denoised Activations

11 Dec 2020ICCV 2021 10arXiv:2012.06440archive 2025-07-28

Sanath Narayan, Hisham Cholakkal, Munawar Hayat, Fahad Shahbaz Khan, Ming-Hsuan Yang, Ling Shao

This work proposes a weakly-supervised temporal action localization framework, called D2-Net, which strives to temporally localize actions using video-level supervision. Our main contribution is the introduction of a novel loss formulation, which jointly enhances the discriminability of latent embeddings and robustness of the output temporal class activations with respect to foreground-background noise caused by weak supervision. The proposed formulation comprises a discriminative and a denoising loss term for enhancing temporal action localization. The discriminative term incorporates a classification loss and utilizes a top-down attention mechanism to enhance the separability of latent foreground-background embeddings. The denoising loss term explicitly addresses the foreground-background noise in class activations by simultaneously maximizing intra-video and inter-video mutual information using a bottom-up attention mechanism. As a result, activations in the foreground regions are emphasized whereas those in the background regions are suppressed, thereby leading to more robust predictions. Comprehensive experiments are performed on multiple benchmarks, including THUMOS14 and ActivityNet1.2. Our D2-Net performs favorably in comparison to the existing methods on all datasets, achieving gains as high as 2.3% in terms of mAP at IoU=0.5 on THUMOS14. Source code is available at https://github.com/naraysa/D2-Net

PaperPDFConference PDFCode

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

Code

naraysa/D2-Net 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 LocalizationDenoisingTemporal Action LocalizationWeakly Supervised Action LocalizationWeakly-supervised Temporal Action Localization

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Weakly Supervised Action Localization ActivityNet-1.2 D2-Net Mean mAP 26 #8 of 19 Archive leaderboard report
Weakly Supervised Action Localization ActivityNet-1.2 D2-Net mAP@0.5 42.3 #8 of 19 Archive leaderboard report
Weakly Supervised Action Localization FineAction D2-Net mAP 3.35 #3 of 4 Archive leaderboard report
Weakly Supervised Action Localization FineAction D2-Net mAP IOU@0.5 6.75 #3 of 4 Archive leaderboard report
Weakly Supervised Action Localization FineAction D2-Net mAP IOU@0.75 3.02 #3 of 4 Archive leaderboard report
Weakly Supervised Action Localization FineAction D2-Net mAP IOU@0.95 0.82 #3 of 4 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 D2-Net mAP@0.1:0.5 51.4 #23 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 D2-Net mAP@0.1:0.7 - #23 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 D2-Net mAP@0.5 35.9 #23 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS’14 D2-Net mAP@0.5 35.9 #4 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.

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