AI Paper Atlas

Research map / Video / Action recognition

Contrastive Learning of Image Representations With Cross-Video Cycle-Consistency

Haiping Wu, Xiaolong Wang

ICCV 2021 · Video › Action recognition · approach: Contrastive video representation learning · 26 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 cross-video cycle loss to intra-image + intra-video loss (ablation)contrastive pretraining (ResNet-18) · OTB-100 trackingprecision: 60.1 -> 65.6Improved
add cross-video cycle loss to intra-image + intra-video loss (ablation)contrastive pretraining (ResNet-50) · OTB-100 trackingprecision: 68.4 -> 69.3Improved
add intra-video invariance to intra-image loss (ablation)contrastive pretraining (ResNet-18) · OTB-100 trackingprecision: 53.7 -> 60.1Improved
add intra-video invariance to intra-image loss (ablation)contrastive pretraining (ResNet-50) · OTB-100 trackingprecision: 47.4 -> 68.4Improved
add intra-video invariance to intra-image loss (ablation)contrastive pretraining (ResNet-18) · ImageNet linear probetop-1 accuracy: 33.0 -> 33.1No clear effect
add cross-video cycle loss to intra-image + intra-video loss (ablation)contrastive pretraining (ResNet-18) · ImageNet linear probetop-1 accuracy: 33.1 -> 34.4Improved
add cross-video cycle loss to intra-image + intra-video loss (ablation)contrastive pretraining (ResNet-50) · ImageNet linear probetop-1 accuracy: 55.1 -> 55.6Improved
add intra-video invariance to intra-image loss (ablation)contrastive pretraining (ResNet-18) · UCF101 action recognitiontop-1 accuracy: 45.1 -> 48.4Improved
add cross-video cycle loss to intra-image + intra-video loss (ablation)contrastive pretraining (ResNet-18) · UCF101 action recognitiontop-1 accuracy: 48.4 -> 50.5Improved
add cross-video cycle loss to intra-image + intra-video loss (ablation)contrastive pretraining (ResNet-50) · UCF101 action recognitiontop-1 accuracy: 66.2 -> 67.1Improved
add intra-video loss to cycle loss alone (ablation)cross-video cycle objective · ImageNet-100 linear probetop-1 accuracy: 45.78 -> 58.50Improved
enlarge neighbor set from 256 to 16384 (ablation)full loss, ResNet-18 · ImageNet-100 linear probetop-1 accuracy: 56.68 -> 58.48Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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