Research map / Detection and segmentation / Weakly supervised detection and segmentation
A Weakly Supervised Fine Label Classifier Enhanced by Coarse Supervision
Fariborz Taherkhani, Hadi Kazemi, Ali Dabouei, Jeremy Dawson, Nasser M. Nasrabadi
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 contrastive loss to jointly trained model (ablation) | DFC-J · ImageNet2010 subset (Scoarse=0.5|S|, Sfine=0.1|S|) | top-1 accuracy (%): 76.23 -> 76.89 | Improved |
| add contrastive loss to jointly trained model (ablation) | DFC-J · ImageNet2010 subset (Sfine=0.5|S|, Scoarse=0.1|S|) | top-1 accuracy (%): 78.63 -> 79.15 | Improved |
| add contrastive loss to separately trained model (ablation) | DFC-S · ImageNet2010 subset (Scoarse=0.5|S|, Sfine=0.1|S|) | top-1 accuracy (%): 72.84 -> 74.32 | Improved |
| train components jointly instead of separately (ablation) | DFC-S-CL · ImageNet2010 subset (Scoarse=0.5|S|, Sfine=0.1|S|) | top-1 accuracy (%): 74.32 -> 76.89 | Improved |
| add low rank and sparse self-expressive constraints (ablation) | Base-J-CL · ImageNet2010 subset (Scoarse=0.5|S|, Sfine=0.1|S|) | top-1 accuracy (%): 69.19 -> 76.89 | Improved |
| add low rank and sparse self-expressive constraints (ablation) | Base-J-CL · ImageNet2010 subset (Scoarse=0.5|S|, Sfine=0.5|S|) | top-1 accuracy (%): 72.64 -> 81.17 | Improved |
| use coarse-labeled images lacking fine labels (ablation) | DFC-J-CL · ImageNet2010 subset (Scoarse=0.5|S|, Sfine=0.1|S|) | top-1 accuracy (%): 73.92 -> 76.89 | Improved |
| increase fine-labeled set from 0.1|S| to 0.5|S| (ablation) | DFC-J-CL · ImageNet2010 subset (Scoarse=0.5|S|) | top-1 accuracy (%): 76.89 -> 78.83 -> 81.17 | Improved |
| increase coarse-labeled set from 0.1|S| to 0.5|S| (ablation) | DFC-J-CL · ImageNet2010 subset (Sfine=0.5|S|) | top-1 accuracy (%): 79.15 -> 79.84 -> 81.17 | Improved |
Closest papers by meaning
- Grafit: Learning Fine-Grained Image Representations With Coarse Labels (ICCV 2021 · similarity 0.88)
- Weakly Supervised Representation Learning With Coarse Labels (ICCV 2021 · similarity 0.87)
- Fine-grained Classes and How to Find Them (ICML 2024 · similarity 0.87)
- Webly-Supervised Fine-Grained Visual Categorization via Deep Domain Adaptation (TPAMI 2016 · similarity 0.87)
- Learning Separable Fine-Grained Representation via Dendrogram Construction from Coarse Labels for Fine-grained Visual Recognition (ICCV 2025 · similarity 0.86)
- Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 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.
