We approach the [VIS task] from a new perspective and propose a one-stage spatial granularity network (SG-Net) (as shown in the above figure). This repo is the implementation of CVPR 2021 paper "SG-Net: Spatial Granularity Network for One-Stage Video Instance Segmentation." [pdf]
Please find detailed steps Here for installation and dataset preparation.
Please find details Here for step-by-step instructions.
Please refer to Here for inference.
- Baseline model code release
- Pretrained weights for baseline model
- Full mode code release
- Pretrained weights for full model
SG-Net is released under the MIT license.
We truely thanksful of the following piror efforts in terms of knowledge contributions and open-source repos.
- BlendMask: Top-down meets bottom-up for instance segmentation (CVPR'20) [paper] [official code]
- Video Instance Segmenation (ICCV'19) [paper] [official code]