uber-research/UPSNet
A unified panoptic segmentation neural network that jointly performs semantic and instance segmentation for scene understanding.
Not currently ranked — collecting fresh signals.
star history
UPSNet is a deep learning architecture for panoptic segmentation that combines semantic and instance segmentation into a single network. The model was introduced in a CVPR 2019 oral paper and is implemented in PyTorch, supporting multi-GPU training via horovod. It provides pretrained models and benchmarks on COCO and Cityscapes datasets, achieving competitive results on standard segmentation metrics.
Frequently asked
- What is uber-research/UPSNet?
- A unified panoptic segmentation neural network that jointly performs semantic and instance segmentation for scene understanding.
- Is UPSNet open source?
- Yes — uber-research/UPSNet is an open-source project tracked on heatdrop.
- What language is UPSNet written in?
- uber-research/UPSNet is primarily written in Python.
- How popular is UPSNet?
- uber-research/UPSNet has 646 stars on GitHub.
- Where can I find UPSNet?
- uber-research/UPSNet is on GitHub at https://github.com/uber-research/UPSNet.