Research map / Label-efficient and robust learning / Recognition and long-tail
GistNet: A Geometric Structure Transfer Network for Long-Tailed Recognition
Bo Liu, Haoxiang Li, Hao Kang, Gang Hua, Nuno Vasconcelos
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 cosine classifier and class-balanced sampling (ablation) | Plain Model · ImageNet-LT val (overall) | accuracy: 25.1 -> 37.6 | Improved |
| add class structure parameters (ablation) | COS+CB · ImageNet-LT val (overall) | accuracy: 37.6 -> 39.5 | Improved |
| add class structure parameters (ablation) | COS+CB · ImageNet-LT val (few-shot) | accuracy: 14.7 -> 14.5 | No clear effect |
| add GIST training (ablation) | COS+CS+CB · ImageNet-LT val (overall) | accuracy: 39.5 -> 43.5 | Improved |
| add GIST training (ablation) | COS+CS+CB · ImageNet-LT val (few-shot) | accuracy: 14.5 -> 21.4 | Improved |
| add class structure parameters (ablation) | COS+GIST · ImageNet-LT val (overall) | accuracy: 40.2 -> 43.5 | Improved |
| use separate centers for random-sampling loss (v_k) rather than shared w_k (ablation) | GistNet · ImageNet-LT val (few-shot) | accuracy: 14.8 -> 21.4 | Improved |
| use separate centers for random-sampling loss (v_k) rather than shared w_k (ablation) | GistNet · ImageNet-LT val (many-shot) | accuracy: 58.2 -> 54.8 | Hurt |
| replace addition g with rotation g (ablation) | GistNet · ImageNet-LT val (overall) | accuracy: 43.5 -> 43.6 | No clear effect |
| replace addition g with MLP g (ablation) | GistNet · ImageNet-LT val (overall) | accuracy: 43.5 -> 43.4 | No clear effect |
| increase structure parameters beyond m=4 (to 8) (ablation) | GistNet · ImageNet-LT val (overall) | accuracy | No clear effect |
| increase structure parameters to m=8 (ablation) | GistNet · ImageNet-LT val (few-shot) | accuracy: one-point gain | Improved |
Closest papers by meaning
- Geometric Prior Guided Feature Representation Learning for Long-Tailed Classification (IJCV 2024 · similarity 0.87)
- Decoupling Representation and Classifier for Long-Tailed Recognition (ICLR 2020 · similarity 0.87)
- BBN: Bilateral-Branch Network With Cumulative Learning for Long-Tailed Visual Recognition (CVPR 2020 · similarity 0.86)
- Long-Tailed Recognition via Weight Balancing (CVPR 2022 · similarity 0.85)
- Long-Tail Recognition via Compositional Knowledge Transfer (CVPR 2022 · similarity 0.85)
- Long-Tailed Learning as Multi-Objective Optimization (AAAI 2024 · similarity 0.84)
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.
