AI Paper Atlas

Research map / Video / Action recognition

AR-Net: Adaptive Frame Resolution for Efficient Action Recognition

Yue Meng, Chung-Ching Lin, Rameswar Panda, Prasanna Sattigeri, Leonid Karlinsky, Aude Oliva, Kate Saenko, Rogerio Feris

ECCV 2020 · Video › Action recognition · approach: Disentangled action recognition · 167 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 frame skipping to the resolution-only action space (ablation)AR-Net (ResNet) · ActivityNet-v1.3mAP: 73.4 -> 73.8Improved
add adaptive resolution to the skipping-only action space (ablation)AR-Net (ResNet) · ActivityNet-v1.3mAP: 72.7 -> 73.8Improved
add efficiency (FLOPS) loss to accuracy loss (ablation)AR-Net (ResNet) · ActivityNet-v1.3GFLOPS/V: 60.06 -> 36.48Improved
add efficiency (FLOPS) loss to accuracy loss (ablation)AR-Net (ResNet) · ActivityNet-v1.3mAP: 74.5 -> 73.8Hurt
add uniform-usage loss regularizer (ablation)AR-Net (ResNet) · ActivityNet-v1.3GFLOPS/V: 36.48 -> 33.47Improved
add uniform-usage loss regularizer (ablation)AR-Net (ResNet) · ActivityNet-v1.3mAP: 73.8 -> 73.8No clear effect
add warm-up stage before joint training (ablation)AR-Net (ResNet) · ActivityNet-v1.3mAP: 67.1 -> 73.3Improved
add fine-tuning stage after joint training (ablation)AR-Net (ResNet) · ActivityNet-v1.3mAP: 73.3 -> 73.8Improved
replace random policy with learned policyAR-Net (ResNet) · ActivityNet-v1.3mAP: 65.0 -> 73.8Improved
replace random policy with learned policyAR-Net (ResNet) · FCVIDmAP: 75.3 -> 81.3Improved
replace uniform full-resolution processing with adaptive policyResNet-50 · ActivityNet-v1.3mAP: 72.5 -> 73.8Improved
replace Multi-Scale baseline with AR-NetResNet · ActivityNet-v1.3GFLOPS/V: 110.43 -> 33.47Improved

Later papers that compared with this method

12 later accepted papers report AR-Net as a baseline in their results tables; 12 of them report a better number on most of the results they share with it.

4 more not listed.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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