AI Paper Atlas

Research map / Video / Action recognition

Towards Efficient Coarse-to-Fine Networks for Action and Gesture Recognition

Niamul Quader, Juwei Lu, Peng Dai, Wei Li

ECCV 2020 · Video › Action recognition · approach: Spatiotemporal CNNs for action recognition · 16 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
extend TSM to C2F ensemble (classroom learning)TSM-ResNet50 · Something-Something V1top-1 accuracy (%): 47.0 -> 48.4Improved
extend TSM to C2F ensemble (classroom learning)TSM-ResNet50 · Something-Something V2top-1 accuracy (%): 61.5 -> 62.4Improved
extend TSM to C2F ensembleTSM-ResNet50 · Jestertop-1 accuracy (%): 96.3 -> 96.5No clear effect
add non-local blocks to TSMTSM-ResNet50 · Something-Something V1top-1 accuracy (%): 47.0 -> 41.3Hurt
add non-local blocks to C2FC,EnC2FC,En · Kinetics-400top-1 accuracy (%): 70.5 -> 71.4Improved
use coarse-exit instead of full ensembleC2FC · Something-Something V2FLOPs and top-1: 62.4 -> 61.8; FLOPs 1.08x -> 0.7xMixed
feed coarse pathways non-downsampled input instead of C2F decomposition (remove decomposition) (ablation)C2FC · Something-Something V1delta top-1: 1.1 -> 1.4Improved
add reparameterized FC layer instead of regular FC (ablation)C2FC · Jestertop-1 accuracy (%): about 0.1 gainImproved
add reparameterized FC layer instead of regular FC (ablation)C2FC · Kinetics-400top-1 accuracy (%): about 0.6 gainImproved
add reparameterized FC layer instead of regular FC (ablation)C2FC · Something-Something V1 and V2top-1 accuracy (%): about 0.3 gainImproved
use additive softmax ensemble of pathways instead of C2F fusion (ablation)C2F · Jestertop-1 accuracy (%): 96.3 -> 95.3Hurt
end-to-end multi-stream loss without C2F fusion design (ablation)C2F · Jestertop-1 accuracy (%): 96.3 -> 96.2No clear effect

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