AI Paper Atlas

Research map / Video / Video segmentation and tracking

Pyramid Constrained Self-Attention Network for Fast Video Salient Object Detection

Yuchao Gu, Lijuan Wang, Ziqin Wang, Yun Liu, Ming-Ming Cheng, Shao-Ping Lu

AAAI 2020 · Video › Video segmentation and tracking · approach: Self-supervised video representation learning · 174 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 PCSA module (ablation)MobileNetV3 static SOD baseline · DAVIS testmax F: .856 -> .880Improved
add non-local block (ablation)MobileNetV3 static SOD baseline · DAVIS testmax F: .856 -> .861Improved
replace non-local with constrained self-attention (r=5) (ablation)MobileNetV3 VSOD · DAVIS testmax F: .861 -> .875Improved
increase CSA window size from 5 to 7 (ablation)single CSA · DAVIS testmax F: .875 -> .873No clear effect
use pyramid of CSAs with multiple window sizes/dilations instead of single CSA (ablation)single CSA r=5 · DAVIS testmax F: .875 -> .880Improved
increase frames per segment T beyond 5 (ablation)PCSA · DAVISmax FHurt
PCSA vs SSAVVSOD · DAVSODmax F: 0.603 -> 0.655Improved
PCSA vs SSAVVSOD · FBMSmax F: 0.865 -> 0.831Hurt

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