Abstract: Conventional video matting outputs one alpha matte for all instances appearing in a video frame so that individual instances are not distinguished. While video instance segmentation provides ...
This project is the official implementation of our accepted NeurIPS 2023 paper BiMatting: Efficient Video Matting via Binarization [PDF]. Created by researchers from Beihang University and ETH Zürich.
Unified Python toolkit for video processing with state-of-the-art AI models. Segment objects, estimate depth, remove backgrounds, and inpaint videos with a single CLI.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback