AI Paper Atlas

Research map / Video / Video segmentation and tracking

Video-Text Representation Learning via Differentiable Weak Temporal Alignment

Dohwan Ko, Joonmyung Choi, Juyeon Ko, Shinyeong Noh, Kyoung-Woon On, Eun-Sol Kim, Hyunwoo J. Kim

CVPR 2022 · Video › Video segmentation and tracking · approach: Self-supervised video representation learning · 23 citations (OpenAlex)

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use VT-TWINS instead of MIL-NCE with S3Dvideo representations · HMDBaccuracy: 53.1 -> 57.9Improved
Use VT-TWINS instead of MIL-NCE with S3Dvideo representations · UCFaccuracy: 82.7 -> 85Improved
Use VT-TWINS instead of the strongest listed baselinevideo-text representations · YouCook2 text-to-video retrievalR@1: 9.6 -> 9.7Improved
Use VT-TWINS instead of MIL-NCEvideo-text representations · MSR-VTT text-to-video retrievalR@1: 8.2 -> 9.4Improved
Use VT-TWINS instead of reproduced MIL-NCEvideo-text representations · YouCook2 video-to-text retrievalR@10: 37.36 -> 40.3Improved
Use VT-TWINS instead of reproduced MIL-NCEvideo-text representations · MSR-VTT video-to-text retrievalR@10: 27.2 -> 29.1Improved
Use VT-TWINS instead of ActBERTaction-step localization representations · CrossTaskCTR: 37.1 -> 40.7Improved
Add temporal augmentation that suppresses semantically strong permutations (ablation)standard-DTW contrastive learning · YouCook2R@10: 8.7 -> 9.6Improved
Use random temporal permutations instead of no temporal augmentation (ablation)standard-DTW contrastive learning · YouCook2R@10: 8.7 -> 5Hurt
Encourage semantically strong temporal permutations instead of no temporal augmentation (ablation)standard-DTW contrastive learning · YouCook2R@10: 8.7 -> 4.9Hurt
Add weak alignment to temporal augmentation A (ablation)standard-DTW contrastive learning · YouCook2R@10: 9.6 -> 10.6Improved
Add local neighborhood smoothing to augmentation A and weak alignment (ablation)standard-DTW contrastive learning · YouCook2R@10: 10.6 -> 12.5Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy