Research map / Detection and segmentation / CNN features
GCoNet+: A Stronger Group Collaborative Co-Salient Object Detector
Peng Zheng, Huazhu Fu, Deng-Ping Fan, Qi Fan, Jie Qin, Yu‐Wing Tai, Chi–Keung Tang, Luc Van Gool
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add network simplification (single 1x1 lateral conv, no multi-stage supervision) (ablation) | GCoNet baseline · CoCA (train DUTS_class) | E-measure max: 0.760 -> 0.752 | Mixed |
| add batch normalization after simplification (ablation) | GCoNet+ (Net-Sim) · CoSal2015 (train DUTS_class) | E-measure max: 0.895 -> 0.912 | Improved |
| add hybrid BCE+IoU loss after simplification (ablation) | GCoNet+ (Net-Sim) · CoCA (train DUTS_class) | E-measure max: 0.752 -> 0.774 | Improved |
| combine network simplification, BN and hybrid loss (ablation) | GCoNet baseline · CoCA (train DUTS_class) | E-measure max: 0.760 -> 0.779 | Improved |
| combine network simplification, BN and hybrid loss (ablation) | GCoNet baseline · CoSOD3k (train DUTS_class) | E-measure max: 0.860 -> 0.882 | Improved |
| combine network simplification, BN and hybrid loss (ablation) | GCoNet baseline · CoSal2015 (train DUTS_class) | E-measure max: 0.888 -> 0.913 | Improved |
| add RACM (recurrent auxiliary classification module) (ablation) | GCoNet+ · CoCA (train DUTS_class) | F-measure max: 0.558 -> 0.570 | Improved |
| add CEM (confidence enhancement module) to RACM (ablation) | GCoNet+ · CoCA (train DUTS_class) | F-measure max: 0.570 -> 0.565 | No clear effect |
| add GST loss to RACM+CEM (ablation) | GCoNet+ · CoCA (train DUTS_class) | E-measure max: 0.779 -> 0.786 | Improved |
| add all of RACM, CEM, GST (ablation) | GCoNet+ · CoCA (train DUTS_class) | MAE: 0.119 -> 0.113 | Improved |
| train on DUTS_class plus COCO-SEG instead of DUTS_class only | GCoNet+ · CoCA | E-measure max: 0.786 -> 0.814 | Improved |
| train on DUTS_class plus COCO-SEG instead of COCO-SEG only | GCoNet+ · CoSal2015 | E-measure max: 0.899 -> 0.924 | Improved |
Later papers that compared with this method
4 later accepted papers report GCoNet+ as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Generalizable Co-Salient Object Detection via Mixed Content-Style Modulation (CVPR 2026 · better on 5 of 6 shared results)
- CONDA: Condensed Deep Association Learning for Co-Salient Object Detection. (ECCV 2024 · better on 5 of 6 shared results)
- Co-Salient Object Detection with Co-Representation Purification (TPAMI 2023 · better on 3 of 3 shared results)
- Advancing Referring Expression Segmentation Beyond Single Image (ICCV 2023 · better on 0 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DeepACG — DeepACG: Co-Saliency Detection via Semantic-Aware Contrast Gromov-Wasserstein Distance (CVPR 2021) · better on 1 of 1 shared result
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
- Group Collaborative Learning for Co-Salient Object Detection (CVPR 2021 · similarity 0.95)
- CoADNet: Collaborative Aggregation-and-Distribution Networks for Co-Salient Object Detection (NeurIPS 2020 · similarity 0.92)
- Re-thinking Co-Salient Object Detection (TPAMI 2021 · similarity 0.90)
- CosalPure: Learning Concept from Group Images for Robust Co-Saliency Detection (CVPR 2024 · similarity 0.89)
- Visual Consensus Prompting for Co-Salient Object Detection (CVPR 2025 · similarity 0.88)
- Towards Learning Group-Equivariant Features for Domain Adaptive 3D Detection (NeurIPS 2024 · similarity 0.86)
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.
