AI Paper Atlas

Research map / Video / Action recognition

Quantifying and Learning Static vs. Dynamic Information in Deep Spatiotemporal Networks

Matthew Kowal, Mennatullah Siam, Md Amirul Islam, Neil D. B. Bruce, Richard P. Wildes, Konstantinos G. Derpanis

TPAMI 2024 · Video › Action recognition · approach: Spatiotemporal CNNs for action recognition · 2 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 fast branch to the slow pathway (SlowFast vs SlowOnly)SlowFast · Kinetics-400 action recognition, stage 2dynamic units in slow pathway: +3.3%Improved
train on SSv2 instead of Kinetics-400 or Diving48 (dataset choice)SlowOnly and FastOnly · Action recognition datasetsdynamic unitsImproved
train with shuffled frames (static-only learning) (ablation)SlowOnly, FastOnly · SSv2 vs Diving48 vs Kinetics-400relative Top-1 accuracyMixed
train on TAO-VOS instead of DAVIS, YouTube-VOS or ImageNetVIDFusionSeg · AVOS fusion layer 5static bias (%): 30.1 -> 25.8 (DAVIS vs TAO-VOS)Improved
train on OVIS instead of YTVISVisTR-R50 · VIS encoder layer sixdynamic biasImproved
remove top-k dynamic units vs static units at inference (ablation)SlowFast · SSv2 top-1 accuracyaccuracy drop: 70% of dynamic units removed: 5% more drop than staticHurt
remove 40% of most static units in fusion layer 4 (ablation)MATNet · MoCAmIoU: 0% when removing only 40%Hurt
add StaticDropout (rates 0.1-0.7) vs standard or no dropout3D-ResNet50 / 18 · SSv2 and Diving48ratio of dynamic unitsImproved
add StaticDropout vs baselines3D-ResNet50 · SSv2 and SomethingElseTop-1 accuracyImproved
raise StaticDropout rate to r=0.7 (ablation)StaticDropout · SSv2 and SomethingElseTop-1 accuracyHurt
StaticDropout (r=0.3) vs standard dropout on rare vs common classes3D-ResNet50 · SomethingElse compositional splitper-class accuracyMixed
use bidirectional (B) instead of unidirectional (U) cross connections with gated fusion (ablation)MATNet/RTNet variants · AVOS fusion layers 4 and 5dynamic biasHurt

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