EasyPortrait -- Face Parsing and Portrait Segmentation Dataset

arXiv:2304.13509v3 Announce Type: replace
Abstract: Recently, video conferencing apps have become functional by accomplishing such computer vision-based features as real-time background removal and face beautification. Limited variability in existing portrait segmentation and face parsing datasets, including head poses, ethnicity, scenes, and occlusions specific to video conferencing, motivated us to create a new dataset, EasyPortrait, for these tasks simultaneously. It contains 40,000 primarily indoor photos repeating video meeting scenarios with 13,705 unique users and fine-grained segmentation masks separated into 9 classes. …

abstract apps arxiv become computer computer vision conferencing cs.cv dataset datasets face features functional head parsing real-time segmentation type video video conferencing vision