idtracker.ai: Tracking all individuals in large collectives of unmarked animals
F. Romero Ferrero, m. g. bergomi, R. Hinz, F. J. Hernandez Heras, G. de Polavieja
Abstract
Our understanding of collective animal behavior is limited by our ability to track each of the individuals. We describe an algorithm and software, idtracker.ai, that extracts from video all trajectories with correct identities at a high accuracy for collectives of up to 100 individuals. It uses two deep networks, one detecting when animals touch or cross and an-other one for animal identification, trained adaptively to conditions and difficulty of the video.
02 /Specification & Classification
Click any tag to browse matching papers- Assay
- Research Focus
- Approach
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- New idtracker.ai: rethinking multi-animal tracking as a representation learning problem to increase accuracy and reduce tracking times
Jordi Torrents, Tiago Costa, Gonzalo G. de Polavieja·bioRxiv·May 30, 2025
- TRex, a fast multi-animal tracking system with markerless identification, and 2D estimation of posture and visual fields
T. Walter, I. D. Couzin·bioRxiv·Oct 14, 2020
- anTraX: high throughput video tracking of color-tagged insects
Asaf Gal, Jonathan Saragosti, Daniel J. C. Kronauer·eLifePeer-reviewed·Nov 19, 2020
- Evidence-based re-identification: Leveraging appearance cues to correct tracking mistakes rather than trying to prevent them
V. Coulombe, Q. Leboulleux, … Benoit Labonte·bioRxiv·Nov 10, 2026
- Tracktor: image-based automated tracking of animal movement and behaviour
V. H. Sridhar, D. G. Roche, S. Gingins·bioRxiv·Sep 9, 2018
- DIPLOMAT: multi-animal tracking with efficient manual editing
Isaac Robinson, George Glidden-Handgis, … Travis J. Wheeler·bioRxiv·Aug 11, 2025