AI Paper Atlas

Research map / Detection and segmentation / CNN features

Co-Salient Object Detection with Co-Representation Purification

Ziyue Zhu, Zhao Zhang, Lin Liu Zheng, Xing Sun, Ming‐Ming Cheng

TPAMI 2023 · Detection and segmentation › CNN features · approach: RGB-D salient object detection · 46 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 SOD head to baseline (ablation)CoRP baseline · CoCAS-measure: 0.560 -> 0.608Improved
add SOD head to baseline (ablation)CoRP baseline · CoSal2015S-measure: 0.699 -> 0.800Improved
add PCS on top of SOD head (ablation)CoRP · CoCAS-measure: 0.608 -> 0.640Improved
add PCS on top of SOD head (ablation)CoRP · CoSOD3kS-measure: 0.753 -> 0.800Improved
add PCS on top of SOD head (ablation)CoRP · CoSal2015S-measure: 0.800 -> 0.869Improved
use PCS instead of using the proxy directly as co-representation (ablation)CoRP · CoCAS-measure: 0.551 -> 0.640Improved
use PCS instead of using the proxy directly as co-representation (ablation)CoRP · CoSOD3kS-measure: 0.744 -> 0.800Improved
use proxy directly as co-representation instead of no proxy (ablation)baseline + SOD · CoCAS-measure: 0.608 -> 0.551Hurt
add RPP recurrent proxy purification (ablation)CoRP · CoCAS-measure: 0.640 -> 0.686Improved
add RPP recurrent proxy purification (ablation)CoRP · CoSOD3kS-measure: 0.800 -> 0.820Improved
increase iterations from T=1 to T=2 (ablation)CoRP · CoCAS-measure: 0.640 -> 0.672Improved
increase iterations from T=3 to T=6 (ablation)CoRP · CoCAS-measure: 0.686 -> 0.693Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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