Title: Quantitative profiling of rodent behavior in individuals and during social interaction
Abstract: Animal behavior is primarily conveyed by movement, and thus a basic understanding of the biology of behavior hinges on the ability to measure how animals move. Movement is also intimately linked to brain health and provides a window into nervous system disorders. In this talk, I will present our recent work to build computer vision tools for 3D movement quantification in individuals and social groups, with a focus on rodents. We trained and benchmarked these tools using large ground-truth 3D pose datasets (Rat 7M and PAIR-R24M) that we acquired via motion capture. I will also summarize how we are using 3D behavioral measurements to improve the sensitivity of drug and disease phenotyping.