Research map / Image generation / Image restoration
Adaptive High-Frequency Transformer for Diverse Wildlife Re-Identification
Chenyue Li, Shuoyi Chen, Mang Ye
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 high-frequency-only training input (pure high-frequency augmentation) (ablation) | ViT ReID baseline · Pigeon | mAP: 70.1 -> 68.4 | Hurt |
| use high-frequency-only training input (ablation) | ViT ReID baseline · Panda | mAP: 40.8 -> 41.8 | Improved |
| add frequency-domain mixed augmentation (ablation) | Pure high-frequency augmentation · Panda | mAP: 41.8 -> 42.7 | Improved |
| add object-aware dynamic selection of high-frequency tokens (ablation) | ViT + FMA · Pigeon | mAP: 70.9 -> 73.6 | Improved |
| add object-aware dynamic selection (ablation) | ViT + FMA · Panda | mAP: 42.7 -> 43.9 | Improved |
| add feature equilibrium loss (ablation) | ViT + FMA + ODS · Panda | mAP: 43.9 -> 44.5 | Improved |
| add feature equilibrium loss (ablation) | ViT + FMA + ODS · Pigeon | mAP: 73.6 -> 73.8 | No clear effect |
| use object-aware selection instead of reproduced PHA (ablation) | PHA (reproduced) · Shark | mAP: 14.8 -> 23.6 | Improved |
| use higher weight for feature equilibrium loss (ablation) | Ours · Panda | mAP | Hurt |
| use full AdaFreq instead of CLIP-ReID in multi-species setting | CLIP-ReID · Elephant (multi-species) | mAP: 20.4 -> 26.6 | Improved |
| AdaFreq versus UniReID in domain generalization on Seal | UniReID · DG Seal | mAP: 28.5 -> 24.0 | Hurt |
Closest papers by meaning
- Toward Re-Identifying Any Animal (NeurIPS 2023 · similarity 0.89)
- Generating the Wild: Individual-Consistent Image-to-Video Generation for Wildlife (NeurIPS 2026 · similarity 0.86)
- Reconstructing Animals and the Wild (CVPR 2025 · similarity 0.84)
- BuckTales: A multi-UAV dataset for multi-object tracking and re-identification of wild antelopes (NeurIPS 2024 · similarity 0.84)
- WildlifeMapper: Aerial Image Analysis for Multi-Species Detection and Identification (CVPR 2024 · similarity 0.84)
- MammAlps: A Multi-view Video Behavior Monitoring Dataset of Wild Mammals in the Swiss Alps (CVPR 2025 · similarity 0.83)
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.
