Research map / Detection and segmentation / Weakly supervised detection and segmentation
Shallow Feature Matters for Weakly Supervised Object Localization
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace addition or concatenation with element-wise multiplication when fusing shallow and deep features (ablation) | CAM generation for WSOL · CUB-200 | GT-known Loc: Addition: 55.76, 55.19, 55.00, 55.10; Concatenation: 55.76, 53.14, 53.40, 53.32; Multiplication: 55.76, 79.05, 88.30, 84 | Helped |
| Add the multiplicative channel-wise attention module (ablation) | MFF-Net · CUB-200 | GT-known Loc: 88.52 -> 92.25 | Helped |
| Add auxiliary loss (ablation) | MFF-Net · CUB-200 | GT-known Loc: 89.45 -> 92.25 | Helped |
| Apply thresholding to Gaussian-enhanced CAM pseudo labels (ablation) | Class-agnostic segmentation · CUB-200 | GT-known Loc: 73.97 -> 96.46 | Helped |
| Add Gaussian enhancement to thresholded CAM pseudo labels (ablation) | Class-agnostic segmentation · CUB-200 | GT-known Loc: 95.41 -> 96.46 | Helped |
| Add the class-agnostic segmentation model (ablation) | SPOL · CUB-200 | GT-known Loc: 92.25 -> 96.46 | Helped |
| Use multiplication fusion across more shallow feature stages rather than conventional addition or concatenation (ablation) | CAM generation for WSOL · CUB-200 | GT-known Loc: Multiplication: 55.76 (Fuse1), 79.05 (Fuse2), 88.30 (Fuse3), 84.04 (Fuse4) | Mixed |
| Use SPOL instead of PSOL | WSOL · CUB-200 | Top-5 Loc: 89.51 -> 93.44 | Helped |
| Use SPOL instead of PSOL | WSOL · CUB-200 | GT-known Loc: 93.01 -> 96.46 | Helped |
| Use SPOL instead of PSOL | WSOL · ImageNet-1K | Top-5 Loc: 65.02 -> 67.15 | Helped |
Closest papers by meaning
- Rethinking the Route Towards Weakly Supervised Object Localization (CVPR 2020 · similarity 0.93)
- Evaluating Weakly Supervised Object Localization Methods Right (CVPR 2020 · similarity 0.93)
- Weakly Supervised Object Localization As Domain Adaption (CVPR 2022 · similarity 0.91)
- Attention-Based Dropout Layer for Weakly Supervised Object Localization (CVPR 2019 · similarity 0.90)
- Unveiling the Potential of Structure Preserving for Weakly Supervised Object Localization (CVPR 2021 · similarity 0.90)
- CAM Back Again: Large Kernel CNNs from a Weakly Supervised Object Localization Perspective (CVPR 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.