AI Paper Atlas

Research map / Video / Action recognition

Learning Self-Similarity in Space and Time As Generalized Motion for Video Action Recognition

Heeseung Kwon, Manjin Kim, Suha Kwak, Minsu Cho

ICCV 2021 · Video › Action recognition · approach: Disentangled action recognition · 53 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 a SELFY blockTSM-R50 (16 frames) · Something-Something V1top-1 accuracy (%): 47.3 -> 54.3Improved
add a SELFY blockTSM-R50 (16 frames) · Something-Something V2top-1 accuracy (%): 61.2 -> 65.7Improved
add a SELFY blockTSM-R50 (16 frames) · Diving-48top-1 accuracy (%): 38.8 -> 41.6Improved
add a SELFY blockTSM-R50 (8 frames) · FineGym Gym288top-1 accuracy (%): 47.9 -> 49.5Improved
add a SELFY blockTSM (16 frames) · Kinetics-400top-1 accuracy (%): 74.7 -> 77.1Improved
lengthen temporal offset range from {1} to {-3..3} (ablation)SELFYNet on TSM-R18 · Something-Something V1 (TSM-R18, 8 frames)top-1 accuracy (%): 47.1 -> 48.6Improved
add temporal range {-3..3} instead of {1} (ablation)SELFYNet on TSN-R18 · Something-Something V1 (TSM-R18, 8 frames)top-1 accuracy (%): 39.7 -> 46.9Improved
use convolution instead of soft-argmax for feature extraction (ablation)SELFYNet on TSM-R18 · Something-Something V1 (TSM-R18, 8 frames)top-1 accuracy (%): 44.0 -> 46.7Improved
use convolution instead of MLP for feature integration (ablation)SELFYNet on TSM-R18 · Something-Something V1 (TSM-R18, 8 frames)top-1 accuracy (%): 47.2 -> 48.4Improved
place SELFY block after res3 rather than res5 (ablation)SELFYNet on TSM-R18 · Something-Something V1 (TSM-R18, 8 frames)top-1 accuracy (%): 42.8 -> 48.4Improved
add STSS features Z to visual features V (ablation)SELFYNet on TSM-R18 · Something-Something V1 (TSM-R18, 8 frames)top-1 accuracy (%): 45.5 -> 48.4Improved
use SELFY instead of self-attention (MHSA) (ablation)TSM-R18 · Something-Something V1 (TSM-R18, 8 frames)top-1 accuracy (%): 44.0 -> 48.4Improved

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