Research map / Vision-language and multimodal / Multimodal LLMs
Region-Aware Multimodal Interleaving for Animal Re-Identification
Yihao Wu, Di Zhao, Wayne Getz, Lingqiao Liu, Gillian Dobbie, Daniel Wilson, Yun Sing Koh
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 region-aware multimodal interleaving vs fine-tuning SigLIP vision encoder only (ablation) | SigLIP · NyalaData | mAP (%): 25.8 -> 35.3 | Improved |
| add region-aware multimodal interleaving vs SigLIP-FT (ablation) | SigLIP · SealID | mAP (%): 20.2 -> 30.1 | Improved |
| add region-aware multimodal interleaving vs SigLIP-FT (ablation) | SigLIP · ATRW | mAP (%): 55.1 -> 63.2 | Improved |
| use localised regions instead of global GloPER patterns (ablation) | RAMI · NyalaData | mAP (%): 34.1 -> 35.3 | Improved |
| use localised regions instead of global patterns (ablation) | RAMI · Giraffes | Rank-1 (%): 88.2 -> 88.0 | No clear effect |
| use localised regions instead of global patterns (ablation) | RAMI · Stoat | mAP (%): 64.8 -> 66.8 | Improved |
| add counterfactual regularisation of the most influential region (ablation) | RAMI · NyalaData | mAP (%): 35.3 -> 34.4 | Hurt |
| add counterfactual regularisation (ablation) | RAMI · Stoat | Rank-1 (%): 91.6 -> 91.9 | No clear effect |
| use RAMI instead of CARE | animal ReID · NyalaData | mAP (%): 16.6 -> 35.3 | Improved |
| use RAMI instead of CARE | animal ReID · Stoat | Rank-1 (%): 91.9 -> 91.6 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CLIP-ReID — CLIP-ReID: Exploiting Vision-Language Model for Image Re-identification without Concrete Text Labels (AAAI 2023) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Toward Re-Identifying Any Animal (NeurIPS 2023 · similarity 0.91)
- Reconstructing Animals and the Wild (CVPR 2025 · similarity 0.87)
- Generating the Wild: Individual-Consistent Image-to-Video Generation for Wildlife (NeurIPS 2026 · similarity 0.86)
- Animal-Bench: Benchmarking Multimodal Video Models for Animal-centric Video Understanding (NeurIPS 2024 · similarity 0.86)
- Meow-Omni 1: A Multimodal Large Language Model for Feline Ethology (NeurIPS 2026 · similarity 0.85)
- SAM 3D Animal: Promptable Animal 3D Reconstruction from Images in the Wild (NeurIPS 2026 · 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.
