lkeab/BCNet
A PyTorch implementation of BCNet, a CVPR 2021 paper on deep occlusion-aware instance segmentation using bilayer decoupled mask heads.
Not currently ranked — collecting fresh signals.
star history
BCNet is a two-stage instance segmentation framework built on detectron2 that explicitly models occlusion by decoupling overlapping objects into bilayer representations. The method improves both anchor-free (FCOS) and anchor-based (Faster R-CNN) detectors by separating object boundary and mask prediction for occluder and occludee within the same RoI. It achieves state-of-the-art results on COCO and KINS benchmarks.
Frequently asked
- What is lkeab/BCNet?
- A PyTorch implementation of BCNet, a CVPR 2021 paper on deep occlusion-aware instance segmentation using bilayer decoupled mask heads.
- Is BCNet open source?
- Yes — lkeab/BCNet is open source, released under the MIT license.
- What language is BCNet written in?
- lkeab/BCNet is primarily written in Python.
- How popular is BCNet?
- lkeab/BCNet has 576 stars on GitHub.
- Where can I find BCNet?
- lkeab/BCNet is on GitHub at https://github.com/lkeab/BCNet.