Perspectives on individual animal identification from biology and computer vision

Abstract

Identifying individual animals is crucial for many biological investigations. In response to some of the limitations of current identification methods, new automated computer vision approaches have emerged with strong performance. Here, we review current advances of computer vision identification techniques to provide both computer scientists and biologists with an overview of the available tools and discuss their applications. We conclude by offering recommendations for starting an animal identification project, illustrate current limitations and propose how they might be addressed in the future.

ElephantBook: A Semi-Automated Human-in-the-Loop System for Elephant Re-Identification

“…Automated re-identification of individual animals using computer vision is an increasingly popular topic, with publications and workshops on the subject at major computer vision conferences [1]. There are several excellent reviews of computer vision for animal re-ID [54,59,68].…”

Seeing biodiversity: perspectives in machine learning for wildlife conservation

“…As an alternative to genebased identification, manual tagging can be used to keep track of individual animals (10,99). Similar to counting and reconstruction (see next section), computer vision recently emerged as a powerful alternative for automatic individual identification (60,94,104,105). The aim is to learn identitybearing features from the appearance of animals.…”

The iWildCam 2021 Competition Dataset

“…A common method for population estimation is mark-recapture, which requires individual animals to be identified and recognized in future imagery [34]. Though strides are being made in visual re-identification for species with strong biometric markings such as zebras [33,37], many species are not visually re-Figure 1. How many pigs are there?…”