DIPLOMAT: multi-animal tracking with efficient manual editing

Abstract Recent advances in computer vision have enabled the development of automated animal behavior observation tools. Several software packages currently exist for concurrently tracking pose in multiple animals; however, existing tools still face challenges in maintaining animal identities across frames and can demand extensive human oversight and editing. Here we report on DIPLOMAT, a D eep learning-based, I dentity- P reserving, L abeled- O bject M ulti- A nimal T racker, which implements automated algorithms improving tolerance to occlusion and continuity of animal identity over a video, further supplemented by an efficient human interface to help eliminate remaining errors. DIPLO-MAT is designed to perform multi-animal tracking by building on the per-frame pose prediction models of two state-of-the-art tools, DeepLabCut and SLEAP. Where other tools immediately take a maximum likelihood estimate from a given video frame, DIPLO-MAT splits the probability fields according to the number of tracked animals and then applies an inference method that takes into account across-frame movement and probabilistic distances between body parts. These independent trace probabilities are then preserved for human editing, enabling multiple body parts to be re-tracked across frames with minimal user action. Testing with a standardized and independently tracked dataset of 3-mouse videos shows DIPLOMAT’s automated components alone can reduce identity swaps by > 75%. DIPLOMAT code and documentation are available at https://diplomattrack.org/ . Significance Statement Tracking multiple animals during social behavior is essential for understanding neural mechanisms underlying social interaction, yet current tools frequently lose individual identities when animals approach/overlap. This identity confusion can require users to spend hours manually correcting tracking errors. DIPLOMAT addresses this bottleneck by integrating neural network and HMM approaches, combining per-frame body-part probabilities with probabilities related to inter-body-part distances and movement through space. We find this reduces identity swap errors without sacrificing other metrics. Moreover, DIPLOMAT stores probability data for use in a manual editing tool that allows for multi-body-part, manual re-tracking and error correction. DIPLOMAT therefore reduces user-time requirements for processing multi-animal recordings, which has the potential to accelerate research into the neural and genetic basis of social behavior.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.