Research map / Detection and segmentation / Weakly supervised detection and segmentation
Normalization Matters in Weakly Supervised Object Localization
Jeesoo Kim, Junsuk Choe, Sangdoo Yun, Nojun Kwak
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 |
|---|---|---|---|
| use IVR normalization instead of min-max | CAM · CUB | MaxBoxAccV2 (mean): 61.13 -> 64.29 | Improved |
| use IVR normalization instead of min-max | HaS · CUB | MaxBoxAccV2 (mean): 60.58 -> 67.85 | Improved |
| use IVR normalization instead of min-max | CutMix · CUB | MaxBoxAccV2 (mean): 60.83 -> 66.80 | Improved |
| use IVR normalization instead of min-max | SPG · CUB | MaxBoxAccV2 (mean): 57.52 -> 61.73 | Improved |
| use IVR normalization instead of min-max | CAM · ImageNet | MaxBoxAccV2 (mean): 62.36 -> 64.18 | Improved |
| use IVR normalization instead of min-max | ADL · ImageNet | MaxBoxAccV2 (mean): 61.65 -> 64.26 | Improved |
| use IVR normalization instead of min-max | CAM · OpenImages | PxAP (mean): 60.01 -> 60.62 | Improved |
| use IVR instead of min-max | ACoL (ResNet) · CUB | MaxBoxAccV2: 66.43 -> 66.33 | No clear effect |
| use max normalization instead of min-max (ablation) | HaS · CUB | MaxBoxAccV2 (mean): 60.58 -> 65.37 | Improved |
| use max normalization instead of min-max (ablation) | CAM · OpenImages | PxAP (mean): 60.01 -> 61.37 | Improved |
| use max normalization instead of min-max (ablation) | ACoL · ImageNet | MaxBoxAccV2 (mean): 61.14 -> 60.95 | No clear effect |
| use PaS instead of min-max (ablation) | CAM · CUB | MaxBoxAccV2 (mean): 61.13 -> 60.23 | Hurt |
Closest papers by meaning
- Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets (TPAMI 2022 · similarity 0.94)
- Evaluating Weakly Supervised Object Localization Methods Right (CVPR 2020 · similarity 0.94)
- Rethinking Class Activation Mapping for Weakly Supervised Object Localization (ECCV 2020 · similarity 0.94)
- Rethinking the Route Towards Weakly Supervised Object Localization (CVPR 2020 · similarity 0.92)
- Background-Aware Classification Activation Map for Weakly Supervised Object Localization (TPAMI 2023 · similarity 0.92)
- Online Refinement of Low-Level Feature Based Activation Map for Weakly Supervised Object Localization (ICCV 2021 · similarity 0.92)
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.
