Skip to main content
. Author manuscript; available in PMC: 2022 Oct 21.
Published in final edited form as: Nat Mach Intell. 2022 Apr 21;4(4):331–340. doi: 10.1038/s42256-022-00477-5

Fig. 4. SIPEC can recognize social interactions of multiple primates and infer their 3D positions using a single camera.

Fig. 4

a) Performance of SIPEC:BehaveNet for individual and social behaviors with respect to ground truth evaluated using grouped 5-fold CV. Behaviors include searching, object interaction and social grooming; while the performance is measured using F1. F1 on shuffled labels is included for comparison. All data is represented by a minimum-to-maximum box-and-whisker plot, showing all points. b) Evaluation of 3D position estimates of primates in home-cage. Black spots mark annotated positions (n=300) while predicted positions are marked as red-hued spots at the end of the solid arrows (color-coded using a red gradient with brighter red indicating higher RMSE of predicted to true position).