Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Object Localization Using Things and Stuff Transfer
Miaojing Shi, Holger Caesar, Vittorio Ferrari
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 TST transfer (APP) to Basic MIL | Basic MIL · ILSVRC-20 (train set) | CorLoc: 39.7 -> 46.7 | Improved |
| add TST transfer (APP) to Basic MIL with objectness | Basic MIL + objectness · ILSVRC-20 (train set) | CorLoc: 47.6 -> 52.7 | Improved |
| replace direct transfer plus MIL with TST (APP) | Basic MIL + objectness · ILSVRC-20 (train set) | CorLoc: 49.1 -> 52.7 | Improved |
| add direct transfer plus MIL to Basic MIL | Basic MIL · ILSVRC-20 (train set) | CorLoc: 39.7 -> 39.5 | No clear effect |
| use direct transfer alone instead of MIL | Basic MIL · ILSVRC-20 (train set) | CorLoc: 39.7 -> 15.0 | Hurt |
| add TST (APP) to Deep MIL with objectness | Deep MIL + objectness · ILSVRC-20 (train set) | CorLoc: 48.4 -> 54.0 | Improved |
| add TST to Deep MIL with objectness | Deep MIL + objectness · ILSVRC-20 (test set) | mAP: 22.3 -> 25.6 | Improved |
| add ILSVRC-dets score to TST (SEM) | Deep MIL + objectness · ILSVRC-20 (test set) | mAP: 25.0 -> 25.9 | Improved |
| add TST (APP) to Deep MIL with objectness | Deep MIL + objectness · COCO-07 | CorLoc: 15.8 -> 18.8 | Improved |
| add TST (SEM) to Deep MIL with objectness and multifolding | Deep MIL + Obj. + MF · PASCAL VOC 2007 | CorLoc: 50.7 -> 59.9 | Improved |
| add TST (SEM) to Deep MIL with objectness and multifolding | Deep MIL + Obj. + MF · PASCAL VOC 2007 | mAP: 28.1 -> 33.8 | Improved |
| add ILSVRC-dets detectors to TST (SEM) | TST (SEM) · PASCAL VOC 2007 | CorLoc: 59.9 -> 60.8 | Improved |
Closest papers by meaning
- Pairwise Similarity Knowledge Transfer for Weakly Supervised Object Localization (ECCV 2020 · similarity 0.89)
- Rethinking the Route Towards Weakly Supervised Object Localization (CVPR 2020 · similarity 0.87)
- Weakly Supervised Object Localization with Multi-Fold Multiple Instance Learning (TPAMI 2016 · similarity 0.87)
- Inter-Image Communication for Weakly Supervised Localization (ECCV 2020 · similarity 0.87)
- Weakly- and Semi-Supervised Panoptic Segmentation (ECCV 2018 · similarity 0.87)
- Weakly Supervised Object Localization via Transformer with Implicit Spatial Calibration (ECCV 2022 · similarity 0.87)
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.
