AI Paper Atlas

Research map / Video / Action recognition

Depthwise Spatio-Temporal STFT Convolutional Neural Networks for Human Action Recognition

Sudhakar Kumawat, Manisha Verma, Yuta Nakashima, Shanmuganathan Raman

TPAMI 2021 · Video › Action recognition · approach: Spatiotemporal CNNs for action recognition · 59 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
use fixed STFT kernel instead of trainable depthwise temporal conv (ablation)T-STFT · Something-Something v1 (64 frames)top-1 accuracy: 46.75 -> 50.63Improved
apply STFT in depthwise fashion instead of ReLPV blocks (ablation)ST-STFT · Something-Something v1 (64 frames)top-1 accuracy: 43.01 -> 48.23Improved
use LeakyReLU instead of ReLU (ablation)ST-STFT · Something-Something v1 (16 frames)top-1 accuracy: 42.62 -> 43.78Improved
use LeakyReLU instead of SELU (ablation)ST-STFT · Something-Something v1 (16 frames)top-1 accuracy: 32.83 -> 43.78Improved
use LeakyReLU instead of ELU (ablation)ST-STFT · Something-Something v1 (16 frames)top-1 accuracy: 32.39 -> 43.78Improved
use temporal-only STFT instead of spatio-temporal STFTSTFT network · Something-Something v1 (64 frames)top-1 accuracy: 48.23 -> 50.63Improved
use temporal STFT instead of trainable depthwise temporal (S-STFT)STFT network · Something-Something v1 (64 frames)top-1 accuracy: 47.12 -> 50.63Improved
increase input frames from 16 to 64T-STFT · Something-Something v1 (from scratch)top-1 accuracy: 46.74 -> 50.63Improved
add Kinetics pre-trainingT-STFT · Something-Something v1 (64 frames)top-1 accuracy: 50.63 -> 52.42Improved
add Kinetics pre-trainingT-STFT · Something-Something v2 (64 frames)top-1 accuracy: 63.05 -> 64.66Improved
add Kinetics pre-trainingT-STFT · Jester (16 frames)top-1 accuracy: 96.85 -> 96.94No clear effect
train from scratch without ImageNet pretrainingT-STFT · Kinetics-400top-1 accuracy: 73.7 -> 61.1Hurt

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