A non-invasive video-based method for individual identification of wildlife using gait dynamics

2026-07-05

AI Summary

This study addresses the long-standing reliance on physical tagging or invasive methods for individual wildlife identification, which lacks scalable, non-invasive solutions suitable for complex natural environments. The authors propose the first fully automated video analysis framework that integrates deep spatiotemporal representation learning with gait analysis. The approach leverages SAM3 to generate high-quality silhouette masks, employs ResNet18 for spatial feature extraction, and utilizes VideoPrism to model temporal dynamics, enabling unsupervised clustering via cosine similarity. Requiring neither visual markings nor physical contact, the method demonstrates strong intra-individual consistency and inter-individual separability across multi-source, multi-species datasets. Quantitative validation through silhouette coefficients and similarity distributions confirms its effectiveness.

Abstract

Gait is a distinctive behavioral characteristic that enables non-invasive individual identification without requiring physical interaction with an animal. While gait-based analysis has been extensively studied in humans, its application to wildlife remains limited due to environmental variability and the lack of scalable identification methods. This paper presents a fully automated, video-based pipeline for wildlife gait analysis and individual identification using deep spatiotemporal representation learning. The proposed pipeline uses the Segment Anything Model 3 (SAM3) to generate high-quality RGB and binary silhouette masks, robustly isolating animals from complex natural backgrounds. Segmented video sequences are processed using a convolutional neural network (ResNet18) for spatial feature extraction and a transformer-based video model (VideoPrism) for temporal motion modeling. Both models are fine-tuned using a classification objective and subsequently used as feature extractors to generate discriminative gait representations. Cosine similarity is then used to compare gait signatures, enabling similarity-based clustering of individuals without reliance on physical markings or invasive tagging. Experiments conducted on multi-source wildlife video data across multiple species demonstrate strong intra-individual consistency and clear inter-individual separation. Quantitative results using cosine similarity distributions and silhouette scores confirm the effectiveness of the proposed method. These findings demonstrate that gait dynamics provide a viable, non-invasive approach for individual identification in wildlife and highlight the potential of video-based deep learning pipelines for scalable ecological monitoring.

Problem

• wildlife identification
• gait dynamics
• non-invasive monitoring
• individual recognition
• video-based analysis

Innovation

• gait-based identification
• video-based wildlife monitoring
• deep spatiotemporal representation
• non-invasive biometrics
• SAM3 segmentation