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. 1. Overview of the SIPEC workflow and modules.

Fig. 1

a) From a given video, instances of animals are segmented with the segmentation network (SIPEC:SegNet), indicated by masked outline as well as bounding boxes. Subsequently, individuals are identified using the identification network (SIPEC:IdNet). For each individual, the pose and behavior can be estimated/classified using the pose estimation network (SIPEC:PoseNet) and the behavioral identification network (SIPEC:BehaveNet), respectively. b) Outcome of SIPEC:SegNet, and SIPEC:IdNet modules are overlaid on a representative videoframe. Time-lapsed positions of individual primates (center of mass) are plotted as circles with respective colors. c) Outputs of SIPEC:SegNet (boxes) and SIPEC:PoseNet (colored dots) on a representative videoframe of mouse open-field data.