KAPAO, a human pose estimator developed by William McNally and John McPhee, as well as co-authors Kanav Vats and Alexander Wong, was accepted for publication at the 2022 European Conference on Computer Vision. Congratulations!
KAPAO is an efficient single-stage multi-person human pose estimation method that models keypoints and poses as objects within a dense anchor-based detection framework. KAPAO simultaneously detects pose objects and keypoint objects and fuses the detections to predict human poses. When not using test-time augmentation, KAPAO is much faster and more accurate than previous single-stage methods like DEKR, HigherHRNet, HigherHRNet + SWAHR, and CenterGroup. The paper can be found here, and the Github repository (550 stars) here.
Dr. McNally completed his PhD in April and is now a senior research engineer at Cleveland Golf.
Monday, July 18, 2022