OpenApePose: a database of annotated ape photographs for pose estimation
This provides a specialized database for researchers studying ape behavior, enabling more reliable pose tracking in naturalistic contexts, though it is incremental as it builds on existing methods for animal tracking.
The researchers tackled the problem of video-based pose tracking for non-human apes by creating OpenApePose, a public dataset of 71,868 annotated photographs with 16 body landmarks across six species, and showed that training a standard deep net on this dataset improves tracking accuracy compared to models trained on monkey or human data.
Because of their close relationship with humans, non-human apes (chimpanzees, bonobos, gorillas, orangutans, and gibbons, including siamangs) are of great scientific interest. The goal of understanding their complex behavior would be greatly advanced by the ability to perform video-based pose tracking. Tracking, however, requires high-quality annotated datasets of ape photographs. Here we present OpenApePose, a new public dataset of 71,868 photographs, annotated with 16 body landmarks, of six ape species in naturalistic contexts. We show that a standard deep net (HRNet-W48) trained on ape photos can reliably track out-of-sample ape photos better than networks trained on monkeys (specifically, the OpenMonkeyPose dataset) and on humans (COCO) can. This trained network can track apes almost as well as the other networks can track their respective taxa, and models trained without one of the six ape species can track the held out species better than the monkey and human models can. Ultimately, the results of our analyses highlight the importance of large specialized databases for animal tracking systems and confirm the utility of our new ape database.