AI Paper Atlas

Research map / Video / Action recognition

Movement Enhancement toward Multi-Scale Video Feature Representation for Temporal Action Detection

Zixuan Zhao, Dongqi Wang, Xu Zhao

ICCV 2023 · Video › Action recognition · approach: Fully supervised temporal action detection · 15 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 movement feature (snippet minus duplicated-frame feature) (ablation)MENet base · ActivityNet-1.3Avg. mAP: 37.0 -> 37.3Improved
add Multi-Relation Enhance Module (ablation)MENet base · ActivityNet-1.3Avg. mAP: 37.0 -> 37.3Improved
combine movement feature and MREM (ablation)MENet · ActivityNet-1.3Avg. mAP: 37.3 -> 37.7Improved
use U-shape module instead of pooling or strided conv downsampling (ablation)SFPN · ActivityNet-1.3Avg. mAP: 37.4 -> 37.7Improved
use multi-layer feature pyramid instead of a single layer (ablation)SFPN · ActivityNet-1.3Avg. mAP: 37.3 -> 37.7Improved
use two-stage training instead of Bias-only sampling (ablation)SFPN ASP training · ActivityNet-1.3Avg. mAP: 29.4 -> 37.7Improved
use both Generalization and Specialization stages instead of one stage (ablation)SFPN ASP training · ActivityNet-1.3Avg. mAP: 37.4 -> 37.7Improved
MENet vs ActionFormer at low tIoU thresholdsMENet · THUMOS-14mAP@0.3: 73.4 -> 70.7Hurt
MENet vs ActionFormer at high tIoU thresholdsMENet · THUMOS-14mAP@0.7: 31.5 -> 34.0Improved

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