AI Paper Atlas

Research map / Video / Action recognition

MorphMLP: An Efficient MLP-Like Backbone for Spatial-Temporal Representation Learning

David Junhao Zhang, Kunchang Li, Yali Wang, Yunpeng Chen, Shashwat Chandra, Yu Qiao, Luoqi Liu, Mike Zheng Shou

ECCV 2022 · Video › Action recognition · approach: Contrastive video representation learning · 50 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
expand chunk length progressively (14/28/28/49) instead of constant 14 (ablation)MorphMLP-S · SSV1top-1 accuracy: 48.7 -> 50.6Improved
expand chunk length progressively instead of global-only chunks (ablation)MorphMLP-S · SSV1top-1 accuracy: 48.6 -> 50.6Improved
expand chunk length progressively instead of constant 3 (ablation)MorphMLP-T · ImageNet-1K (100 epochs)top-1 accuracy: 78.2 -> 79.6Improved
expand chunk length progressively (ablation)MorphMLP-T · ADE20K (ImageNet-pretrained, 100 epochs)mIoU: 41.0 -> 42.6Improved
use sequential temporal-then-spatial instead of parallel (ablation)MorphMLP-S · SSV1top-1 accuracy: 49.2 -> 50.6Improved
use skip residual instead of standard residual (T+S) (ablation)MorphMLP-S · SSV1top-1 accuracy: 49.8 -> 50.6Improved
put spatial before temporal with skip residual (ablation)MorphMLP-S · SSV1top-1 accuracy: 50.6 -> 31.7Hurt
replace MorphFC_s with 3x3 conv (ablation)MorphMLP-T · ImageNet-1K (100 epochs)top-1 accuracy: 79.6 -> 77.3Hurt
replace MorphFC_s with 1D group conv of same kernel sizes (ablation)MorphMLP-T · ImageNet-1K (100 epochs)top-1 accuracy: 79.6 -> 79.0Hurt
replace MorphFC_t with 3x1x1 temporal conv (ablation)MorphMLP-S · SSV1top-1 accuracy: 50.6 -> 47.9Hurt
add channel FC to horizontal and vertical FC (ablation)MorphMLP-S · SSV1top-1 accuracy: 49.4 -> 50.6Improved
add weighted sum after three directions (ablation)MorphMLP-T · ImageNet-1K (100 epochs)top-1 accuracy: 79.3 -> 79.6Improved

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