Spaces:
Sleeping
Sleeping
| title: MCA Classifier | |
| emoji: 🐠 | |
| colorFrom: indigo | |
| colorTo: purple | |
| sdk: gradio | |
| sdk_version: 5.37.0 | |
| app_file: app.py | |
| pinned: false | |
| license: cc-by-4.0 | |
| short_description: Multi-Crop Attention-based Classifier | |
| <div align="center"> | |
| <h1>Paying Attention to Other Animal Detections Improves Camera Trap Classification Models</h1> | |
| <a href="https://www.biorxiv.org/content/10.1101/2025.07.15.664849v1"><img src="https://img.shields.io/badge/bioRxiv-2025.07.15.664849-b31b1b" alt="arXiv"></a> | |
| <a href='https://huggingface.co/spaces/gdussert/MCA_Classifier'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Demo-blue'></a> | |
| **[LBBE](https://lbbe.univ-lyon1.fr/fr)**; **[CEFE](https://www.cefe.cnrs.fr/fr/)**; **[LECA](https://leca.osug.fr/)** | |
| [Gaspard Dussert](https://gdussert.github.io/), [Vincent Miele](https://vmiele.gitlab.io/), [Simon Chamaillé-Jammes](https://simonchamaillejammes.mystrikingly.com/), [Stéphane Dray](https://sdray.github.io/) | |
| </div> | |
| ```bibtex | |
| @misc{dussert2025attention, | |
| author = {Dussert, Gaspard and Dray, Stéphane and Chamaillé-Jammes, Simon and Miele, Vincent}, | |
| title = {Paying Attention to Other Animal Detections Improves Camera Trap Classification Models}, | |
| year = {2025}, | |
| doi = {10.1101/2025.07.15.664849}, | |
| publisher = {bioRxiv}, | |
| type = {preprint}, | |
| } | |
| ``` | |
| ## Overview | |
| Multi-Crop Attention-based Classifier uses self-attention to improve classification by levaring the context of other detection within the camera trap sequence. | |
| GitHub repo is accessible here : [https://github.com/gdussert/MCA_Classifier](https://github.com/gdussert/MCA_Classifier) | |
| Note: Due to the 1GB storage constraint on Hugging Face, the crop_classifier has been converted to float16. | |
|  | |
|  | |
| ## Acknowledgements | |
| This work was granted access to the HPC resources of IDRIS under the allocation 2022-AD010113729 made by GENCI. | |
| Thanks to [LILA BC](https://lila.science/) for hosting the two datasets, [Snapshot Safari 2024 Expansion](https://lila.science/datasets/snapshot-safari-2024-expansion/) and [Snapshot Serengeti](https://lila.science/datasets/snapshot-serengeti), and for all those who contributed to it. | |
| Thanks [MegaDetector](https://github.com/agentmorris/MegaDetector) and [PytorchWildlife](https://github.com/microsoft/CameraTraps): MDv5a was used for preprocessing and MDV6-Ultralytics-YoloV10-Extra is used by the Gradio demo. | |
| Thanks to the amazing [timm](https://github.com/huggingface/pytorch-image-models) repository. | |