AI Paper Atlas

Research map / Video / Action recognition

Learning Temporal Dynamics From Cycles in Narrated Video

Dave Epstein, Jiajun Wu, Cordelia Schmid, Chen Sun

ICCV 2021 · Video › Action recognition · approach: Contrastive video representation learning · 12 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
add a temporal constraint (max-index plus similarity loss) (ablation)MMCC · Recipes testcycle percentile rank: 92.6 -> 93.6Improved
add any temporal constraint against collapse (ablation)MMCC · Recipes testtraining stability: None -> 93.6Improved
sample start nodes by cross-modal concreteness instead of randomly (ablation)MMCC · Recipes testcycle percentile rank: 88.7 -> 93.6Improved
fine-tune input embeddings instead of freezing cross-modal features (ablation)MMCC · Recipes testcycle percentile rank: 67.5 -> 93.6Improved
fine-tune input embeddings instead of freezing cross-modal features (ablation)MMCC · Recipes testcross-modal percentile rank: 76.8 -> 75.7Hurt
cycle within modalities instead of across modalities (ablation)MMCC · Recipes testcycle percentile rank: 85.0 -> 93.6Improved
mix unimodal and multimodal states (p_unimodal 0.5) instead of unimodal only (ablation)MMCC · Recipes testcycle percentile rank: 89.8 -> 93.6Improved
use p_unimodal 0 (always multimodal) instead of 0.5 (ablation)MMCC · Recipes testcycle percentile rank: 93.6 -> 96.5Improved
add the cross-modal loss during co-training (ablation)MMCC · Recipes testcycle accuracy: 7% decrease without itImproved
add the visual similarity loss (ablation)MMCC · Recipes testcycle rank: -1% without itImproved
use MMCC instead of TAP for future predictionTAP · CrossTask future action anticipationrecall@10: 27.9 -> 33.8Improved
use MMCC instead of TAP for unshufflingTAP · Unshuffling image collectionsKendall's tau: 0.3344 -> 0.3632Improved

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