TuSimple/mx-maskrcnn
An MXNet implementation of Mask R-CNN for instance segmentation, benchmarked on Cityscapes dataset.
Not currently ranked — collecting fresh signals.
star history
This repository provides an MXNet implementation of Mask R-CNN, a deep learning model that performs object detection and instance segmentation simultaneously. The implementation includes training code, custom ROIAlign operators, and benchmarks on the Cityscapes dataset showing results around 26.9 mAP on test set. Users need to build MXNet from source with custom operators and prepare the Cityscapes dataset to train the model.
Frequently asked
- What is TuSimple/mx-maskrcnn?
- An MXNet implementation of Mask R-CNN for instance segmentation, benchmarked on Cityscapes dataset.
- Is mx-maskrcnn open source?
- Yes — TuSimple/mx-maskrcnn is open source, released under the Apache-2.0 license.
- What language is mx-maskrcnn written in?
- TuSimple/mx-maskrcnn is primarily written in Python.
- How popular is mx-maskrcnn?
- TuSimple/mx-maskrcnn has 1.8k stars on GitHub.
- Where can I find mx-maskrcnn?
- TuSimple/mx-maskrcnn is on GitHub at https://github.com/TuSimple/mx-maskrcnn.