AI Paper Atlas

Research map / Video / Video segmentation and tracking

DualFormer: Local-Global Stratified Transformer for Efficient Video Recognition

Yuxuan Liang, Pan Zhou, Roger Zimmermann, Shuicheng Yan

ECCV 2022 · Video › Video segmentation and tracking · approach: Self-supervised video representation learning · 31 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 GP-MSA to LW-MSA in all stages (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 78.4 -> 79.5Improved
add LW-MSA to GP-MSA in all stages (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 77.6 -> 79.5Improved
add GP-MSA to early stages rather than late stages (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 78.8 -> 79.3Improved
use two pyramid scales instead of only (4,4,4) (ablation)DualFormer-T GP-MSA · Kinetics-400top-1 accuracy: 78.4 -> 79.5Improved
use two pyramid scales instead of only (8,7,7) (ablation)DualFormer-T GP-MSA · Kinetics-400top-1 accuracy: 79.3 -> 79.5Improved
enlarge LW-MSA window from 4x7x7 to 8x7x7 (32x2 input) (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 79.1 -> 79.5 (FLOPs 224 -> 240)Improved
enlarge LW-MSA window from 8x7x7 to 16x7x7 (32x2 input) (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 79.5 -> 79.7 (FLOPs 240 -> 272)Improved
use 32 input frames instead of 16 (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 78.4 -> 79.5Improved
use DWConv instead of AvgPool for pyramid downsampling (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 78.7 -> 79.5Improved
use DWConv instead of standard Conv for downsampling (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 79.5 -> 79.5 (Param 27.6 -> 21.8)No clear effect
add DWConv-based PEG (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 78.9 -> 79.5Improved
use PEG instead of absolute position encoding (ablation)DualFormer-T · Kinetics-400top-1 accuracy: 79.2 -> 79.5Improved

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