youngwanLEE/CenterMask
CenterMask is a real-time anchor-free instance segmentation model using a spatial attention-guided mask branch with VoVNetV2 backbone.
Not currently ranked — collecting fresh signals.
star history
This repository implements CenterMask, an anchor-free one-stage instance segmentation approach built on FCOS object detection. It adds a novel SAG-Mask branch that predicts segmentation masks using spatial attention maps to focus on informative pixels. The project includes both the full CenterMask model and a lightweight CenterMask-Lite variant designed for real-time inference, achieving state-of-the-art results faster than previous methods like Mask R-CNN and YOLACT.
Frequently asked
- What is youngwanLEE/CenterMask?
- CenterMask is a real-time anchor-free instance segmentation model using a spatial attention-guided mask branch with VoVNetV2 backbone.
- Is CenterMask open source?
- Yes — youngwanLEE/CenterMask is an open-source project tracked on heatdrop.
- What language is CenterMask written in?
- youngwanLEE/CenterMask is primarily written in Python.
- How popular is CenterMask?
- youngwanLEE/CenterMask has 782 stars on GitHub.
- Where can I find CenterMask?
- youngwanLEE/CenterMask is on GitHub at https://github.com/youngwanLEE/CenterMask.