AI Paper Atlas

Research map / Video / Action recognition

Efficient Few-Shot Action Recognition via Multi-Level Post-Reasoning

Cong Wu, Xiao-Jun Wu, Linze Li, Tianyang Xu, Zhenhua Feng, Josef Kittler

ECCV 2024 · Video › Action recognition · approach: Disentangled action recognition · 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 post-reasoning on cached frozen CLIP features (vs FC layer only) (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 41.0 -> 51.6Improved
add multi-level (global, patch, frame) representations (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 51.6 -> 57.1Improved
add skip-fusion over cached multi-stage features (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 55.8 -> 57.1Improved
add interactive spatiotemporal reasoning (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 54.1 -> 57.1Improved
add text-guided temporal reasoning on support branch (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 55.5 -> 57.1Improved
add text-visual matching alongside support-query (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 56.1 -> 57.1Improved
add parallel temporal attention alongside spatial in skip-fusion (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 54.4 -> 57.1Improved
use selective Top-K mechanism on support-query matching (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 57.1 -> 56.5Hurt
use selective mechanism on text-visual matching (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 56.4 -> 57.1Improved
increase CLIP backbone from ViT-B/32 to ViT-B/16 (ablation)EMP-Net · SSv2-Small 1-shotaccuracy (%): 54.4 -> 57.1Improved
use cached features from layers {4,8,12} instead of {6,12} or {3,6,9,12} (ablation)EMP-Net skip-fusion · SSv2-Small 1-shotaccuracy (%): 55.4 / 56.2 -> 57.1Improved
replace trainable CLIP-FSAR with EMP-Net on frozen CLIPCLIP-FSAR · SSv2-Small 1-shot, 8 framesaccuracy (%): 50.1 -> 57.1Improved

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