Research map / Vision-language and multimodal / Human and embodied modeling
The Curious Layperson: Fine-Grained Image Recognition Without Expert Labels
Subhabrata Choudhury, Iro Laina, Christian Rupprecht, Andrea Vedaldi
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 score distribution regularizer R(B) and neutral class (ablation) | FGSM w/ ensemble · CUB-200 | top-1 accuracy: 5.9 -> 7.9 | Improved |
| add score distribution regularizer R(B) (2-class) (ablation) | FGSM w/ ensemble · CUB-200 | top-1 accuracy: 5.9 -> 7.4 | Improved |
| add neutral class on top of regularizer (ablation) | FGSM + R [2-cls] · CUB-200 | top-1 accuracy: 7.4 -> 7.9 | Improved |
| add neutral class on top of regularizer (ablation) | FGSM + R [2-cls] · CUB-200 | mean rank (lower better): 29.9 -> 31.9 | Hurt |
| replace cosine similarity with learned FGSM matching head (ablation) | FGSM · CUB-200 | top-1 accuracy: 4.5 -> 5.9 | Improved |
| use SAT+AoANet caption ensemble instead of SAT alone (ablation) | FGSM · CUB-200 | top-1 accuracy: 4.3 -> 5.9 | Improved |
| use ensemble instead of AoANet alone (ablation) | FGSM · CUB-200 | top-5 accuracy: 20.8 -> 20.0 | Mixed |
| use ground-truth human descriptions instead of generated captions | FGSM · CUB-200 | top-1 accuracy: 5.9 -> 11.9 | Improved |
| use FGSM Siamese scoring instead of cross-encoder RoBERTa | document scoring · CUB-200 | inference time per image: 347 sec -> 0.55 sec | Improved |
| use captions as intermediary instead of DSCMR trained on image-document pairs | ours · CUB-200 (ZSL split) | top-1 accuracy: 13.5 -> 20.9 | Improved |
| remove class names from documents | CLIP · CUB-200 (ZSL split) | retrieval accuracy | Hurt |
Closest papers by meaning
- Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 2024 · similarity 0.90)
- The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition (ECCV 2016 · similarity 0.89)
- Analyzing and Boosting the Power of Fine-Grained Visual Recognition for Multi-modal Large Language Models (ICLR 2025 · similarity 0.86)
- INQUIRE: A Natural World Text-to-Image Retrieval Benchmark (NeurIPS 2024 · similarity 0.85)
- The Devil is in the Fine-Grained Details: Evaluating Open-Vocabulary Object Detectors for Fine-Grained Understanding (CVPR 2024 · similarity 0.85)
- MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions (ICML 2024 · similarity 0.85)
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.
