Message Passing Framework for Vision Prediction Stability in Human Robot Interaction
To stabilise vision prediction results used in HRI, we present a message passing framework (MPF) that uses the memory of individual modules to correct each other's outputs. We make our code publicly available so that other HRI researchers can potentially benefit from more stabilised outputs from their vision modules. This MPF includes pose estimation, object detection, hand-object detection, and gaze estimation.
Paper: “Message Passing Framework for Vision Prediction Stability in Human Robot Interaction”, Jang and Demiris, ICRA 2022 (forthcoming).
Github repository: https://github.com/youngkyoonjang/MessagePassingFramework
Paper website: https://sites.google.com/view/mpf-hri