AI Paper Atlas

Research map / Video / Action recognition

Approximated Bilinear Modules for Temporal Modeling

Xinqi Zhu, Chang Xu, Langwen Hui, Cewu Lu, Dacheng Tao

ICCV 2019 · Video › Action recognition · approach: Spatiotemporal CNNs for action recognition · 33 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
feed concatenated frames to both ABM entries (ABM-C) instead of adjacent frames separately (ABM-S) (ablation)ABM-top L=3 · Something-Something v1 valtop-1 accuracy: 30.31 -> 38.32Improved
stack 3 ABM-C layers instead of 1 (ablation)ABM-C-top · Something-Something v1 valtop-1 accuracy: 35.49 -> 38.32Improved
stack 3 ABM-S layers instead of 1 (ablation)ABM-S-top · Something-Something v1 valtop-1 accuracy: 29.94 -> 30.31No clear effect
sample snippets of 3 frames instead of single frame per segment (ablation)ABM-C-top L=3 · Something-Something v1 valtop-1 accuracy: 38.32 -> 42.35Improved
implant ABMs into CNN layers instead of placing on top (ablation)ABM-C 8x3 · Something-Something v1 valtop-1 accuracy: 42.35 -> 44.14Improved
increase dynamic fraction beta from 1/4 to 1/2 to 1 (ablation)ABM-A-in 16x3 · Something-Something v1 valtop-1 accuracy: 44.89 -> 45.67 -> 46.08Improved
add shifting inference (ST=3) (ablation)ABM-A / ABM-C · Something-Something v1 valtop-1 accuracy: 44.89 -> 45.56; 45.67 -> 46.16; 46.08 -> 46.81Improved
use ABM-C-top instead of TRN as post-CNN temporal model8-frame ResNet backbone · Something-Something v1 valtop-1 accuracy: 31.68 -> 38.32Improved
add snippet sampling and shifting inference (ablation)ABM-C-in ResNet-50 · Something-Something v1 valtop-1 accuracy: 47.45 -> 49.83Improved
ensemble ABM-C-in and ABM-A-inResNet-34 ABM · Something-Something v1 valtop-1 accuracy: 46.81 -> 49.02Improved
add optical flow stream with 1:1 weighted fusionABM-C-in · Something-Something v1 valtop-1 accuracy: 46.81 -> 50.09Improved
add ABM-C module to I3D backboneI3D · Kineticstop-1 accuracy: 71.1 -> 72.6Improved

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