Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Object Detection With Segmentation Collaboration
Xiaoyan Li, Meina Kan, Shiguang Shan, Xilin Chen
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 a segmentation branch as plain multi-task without collaboration (ablation) | OICR detection baseline · VOC 2007 test | mAP: 41.2 -> 41.3 | No clear effect |
| add one-way segmentation-to-detection guidance (ablation) | Det + Seg · VOC 2007 test | mAP: 41.3 -> 36.8 | Hurt |
| add detection-to-segmentation guidance (two-way collaboration) (ablation) | Det + Seg + Seg->Det · VOC 2007 test | mAP: 36.8 -> 48.3 | Improved |
| add full segmentation-detection collaboration loop (ablation) | OICR baseline · VOC 2007 test | mAP: 41.2 -> 48.3 | Improved |
| add Fast RCNN retraining on top detections (multi-stage) (ablation) | SDCN · VOC 2007 test | mAP: 48.3 -> 51.0 | Improved |
| replace OICR with SDCN | OICR-VGG16 · VOC 2007 trainval | CorLoc: 60.6 -> 66.8 | Improved |
| replace OICR with SDCN | OICR-VGG16 · VOC 2012 | mAP: 37.9 -> 43.5 | Improved |
| add Fast RCNN retraining | SDCN · VOC 2012 | mAP: 43.5 -> 46.7 | Improved |
| vary loss balance weights above 0.1 (ablation) | SDCN · VOC 2007 test | mAP | No clear effect |
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.
- TS2C — TS2C: Tight Box Mining with Surrounding Segmentation Context for Weakly Supervised Object Detection (ECCV 2018) · better on 3 of 3 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
- Generative Adversarial Learning Towards Fast Weakly Supervised Detection (CVPR 2018 · similarity 0.91)
- Towards Precise End-to-End Weakly Supervised Object Detection Network (ICCV 2019 · similarity 0.89)
- Weakly Supervised Object Localization and Detection: A Survey (TPAMI 2021 · similarity 0.89)
- Multi-Evidence Filtering and Fusion for Multi-Label Classification, Object Detection and Semantic Segmentation Based on Weakly Supervised Learning (CVPR 2018 · similarity 0.89)
- Weakly Supervised Deep Detection Networks (CVPR 2016 · similarity 0.89)
- Cyclic Guidance for Weakly Supervised Joint Detection and Segmentation (CVPR 2019 · similarity 0.89)
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.
