Wildlife Target Re-Identification Using Self-supervised Learning in Non-Urban Settings

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Muthivhi, Mufhumudzi, van Zyl, Terence L.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916824289902592
author Muthivhi, Mufhumudzi
van Zyl, Terence L.
author_facet Muthivhi, Mufhumudzi
van Zyl, Terence L.
contents Wildlife re-identification aims to match individuals of the same species across different observations. Current state-of-the-art (SOTA) models rely on class labels to train supervised models for individual classification. This dependence on annotated data has driven the curation of numerous large-scale wildlife datasets. This study investigates self-supervised learning Self-Supervised Learning (SSL) for wildlife re-identification. We automatically extract two distinct views of an individual using temporal image pairs from camera trap data without supervision. The image pairs train a self-supervised model from a potentially endless stream of video data. We evaluate the learnt representations against supervised features on open-world scenarios and transfer learning in various wildlife downstream tasks. The analysis of the experimental results shows that self-supervised models are more robust even with limited data. Moreover, self-supervised features outperform supervision across all downstream tasks. The code is available here https://github.com/pxpana/SSLWildlife.
format Preprint
id arxiv_https___arxiv_org_abs_2507_02403
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Wildlife Target Re-Identification Using Self-supervised Learning in Non-Urban Settings
Muthivhi, Mufhumudzi
van Zyl, Terence L.
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Wildlife re-identification aims to match individuals of the same species across different observations. Current state-of-the-art (SOTA) models rely on class labels to train supervised models for individual classification. This dependence on annotated data has driven the curation of numerous large-scale wildlife datasets. This study investigates self-supervised learning Self-Supervised Learning (SSL) for wildlife re-identification. We automatically extract two distinct views of an individual using temporal image pairs from camera trap data without supervision. The image pairs train a self-supervised model from a potentially endless stream of video data. We evaluate the learnt representations against supervised features on open-world scenarios and transfer learning in various wildlife downstream tasks. The analysis of the experimental results shows that self-supervised models are more robust even with limited data. Moreover, self-supervised features outperform supervision across all downstream tasks. The code is available here https://github.com/pxpana/SSLWildlife.
title Wildlife Target Re-Identification Using Self-supervised Learning in Non-Urban Settings
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2507.02403