Disentangling rodent behaviors to improve automated behavior recognition

Frontiers in Neuroscience

Neuroscience

Marcel A. J. Van Gerven¹  Elsbeth A. Van Dam²  Lucas P. J. J. Noldus³ 

[1] Department of Artificial Intelligence, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, Netherlands;Department of Artificial Intelligence, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, Netherlands;Noldus Information Technology BV, Wageningen, Netherlands;Noldus Information Technology BV, Wageningen, Netherlands;Department of Biophysics, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, Netherlands;

关键词: action recognition;    deep learning;    continuous video analysis;    behavior recognition;    rodent behavior;   

DOI  :  10.3389/fnins.2023.1198209

 received in 2023-03-31, accepted in 2023-06-12,  发布年份 2023

来源: Frontiers

【 摘 要 】

Automated observation and analysis of behavior is important to facilitate progress in many fields of science. Recent developments in deep learning have enabled progress in object detection and tracking, but rodent behavior recognition struggles to exceed 75–80% accuracy for ethologically relevant behaviors. We investigate the main reasons why and distinguish three aspects of behavior dynamics that are difficult to automate. We isolate these aspects in an artificial dataset and reproduce effects with the state-of-the-art behavior recognition models. Having an endless amount of labeled training data with minimal input noise and representative dynamics will enable research to optimize behavior recognition architectures and get closer to human-like recognition performance for behaviors with challenging dynamics.

【 授权许可】

Unknown   
Copyright © 2023 Van Dam, Noldus and Van Gerven.