Papers › Negative Sample Matters: A Renaissance of Metric Learning for Temporal Grounding

Negative Sample Matters: A Renaissance of Metric Learning for Temporal Grounding

10 Sep 2021arXiv:2109.04872archive 2025-07-28

Zhenzhi Wang, LiMin Wang, Tao Wu, TianHao Li, Gangshan Wu

Temporal grounding aims to localize a video moment which is semantically aligned with a given natural language query. Existing methods typically apply a detection or regression pipeline on the fused representation with the research focus on designing complicated prediction heads or fusion strategies. Instead, from a perspective on temporal grounding as a metric-learning problem, we present a Mutual Matching Network (MMN), to directly model the similarity between language queries and video moments in a joint embedding space. This new metric-learning framework enables fully exploiting negative samples from two new aspects: constructing negative cross-modal pairs in a mutual matching scheme and mining negative pairs across different videos. These new negative samples could enhance the joint representation learning of two modalities via cross-modal mutual matching to maximize their mutual information. Experiments show that our MMN achieves highly competitive performance compared with the state-of-the-art methods on four video grounding benchmarks. Based on MMN, we present a winner solution for the HC-STVG challenge of the 3rd PIC workshop. This suggests that metric learning is still a promising method for temporal grounding via capturing the essential cross-modal correlation in a joint embedding space. Code is available at https://github.com/MCG-NJU/MMN.

PaperPDFCode

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

Code

mcg-nju/mmn officialmentioned in papermentioned on GitHubpytorchMIT report
aim3-ruc/youmakeup_challenge2022 mentioned 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

Metric LearningRepresentation LearningTemporal Sentence GroundingVideo Grounding

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Temporal Sentence Grounding Charades-STA MMN (Full, MViT-K400-Pretrain-feature, evaluated by AdaFocus) R1@0.5 55.2 #4 of 13 Archive leaderboard report
Temporal Sentence Grounding Charades-STA MMN (Full, MViT-K400-Pretrain-feature, evaluated by AdaFocus) R1@0.7 32.2 #4 of 13 Archive leaderboard report
Temporal Sentence Grounding Charades-STA MMN (Full, MViT-K400-Pretrain-feature, evaluated by AdaFocus) R5@0.5 88.3 #4 of 13 Archive leaderboard report
Temporal Sentence Grounding Charades-STA MMN (Full, MViT-K400-Pretrain-feature, evaluated by AdaFocus) R5@0.7 62.7 #4 of 13 Archive leaderboard report
Temporal Sentence Grounding Charades-STA MMN (Full, I3D-K400-Pretrain-feature, evaluated by AdaFocus) R1@0.5 49.4 #5 of 13 Archive leaderboard report
Temporal Sentence Grounding Charades-STA MMN (Full, I3D-K400-Pretrain-feature, evaluated by AdaFocus) R1@0.7 29.8 #5 of 13 Archive leaderboard report
Temporal Sentence Grounding Charades-STA MMN (Full, I3D-K400-Pretrain-feature, evaluated by AdaFocus) R5@0.5 85.8 #5 of 13 Archive leaderboard report
Temporal Sentence Grounding Charades-STA MMN (Full, I3D-K400-Pretrain-feature, evaluated by AdaFocus) R5@0.7 60.5 #5 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