Authors: Amanda Duarte, et al. Venue: CVPR 2021
The naming convention "BFH" typically refers to keypoints, which are high-resolution 2D or 3D skeletal data extracted from RGB videos. In the How2Sign dataset: BFH03_03.mp4
The video file is a specific clip from the How2Sign dataset , a large-scale multimodal collection of American Sign Language (ASL) videos designed for computer vision research. Primary Research Paper Authors: Amanda Duarte, et al
Other related research, such as the , also uses a BFH subset for evaluating the accuracy of 3D human meshes. Primary Research Paper Other related research, such as
Related Research at BFH (Bern University of Applied Sciences)
are specialized subsets that include these detailed pose estimations for use in tasks like sign language translation and human pose regression.
The foundational paper for this dataset, which describes its collection, annotation, and metadata, is: