# TuSimple/mx-maskrcnn

An MXNet implementation of Mask R-CNN

Repository: https://github.com/TuSimple/mx-maskrcnn
Canonical: https://ross.abutalabs.com/products/mx-maskrcnn
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2018-02-28T02:51:13+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3235, "days_push": 3108, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1753, forks 542 (observed 2026-08-28T04:05:31.830588+00:00)

## What it is
An MXNet implementation of the Mask R-CNN instance segmentation model, built on top of the mx-rcnn Faster R-CNN codebase. It includes training, evaluation, and demo scripts targeting the Cityscapes dataset with a ResNet-50-FPN backbone.

## Use cases
- train a Mask R-CNN model on Cityscapes with MXNet
- run instance segmentation on single images
- evaluate instance segmentation mAP on Cityscapes val/test
- republish ROIAlign operator in a custom MXNet build
- use a pretrained Mask R-CNN model for demo inference

## When to choose
- you need Mask R-CNN specifically on MXNet rather than PyTorch or TensorFlow
- you are working with Cityscapes-style urban scene segmentation
- you want a reference implementation based on mx-rcnn

## When to avoid
- you need COCO training, which was never completed
- you want an actively maintained implementation
- you are on Python 3 or modern MXNet versions, since it targets Python 2.7 and a patched MXNet build

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, machine-learning, deep-learning, image-processing
- platform: python
- tags: mask-rcnn, instance-segmentation, mxnet, object-detection, cityscapes, coco, linux, gpu

## Member repositories
- TuSimple/mx-maskrcnn (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:31.830588+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:28:10.058511+00:00, confidence not recorded.
  - readme: https://github.com/TuSimple/mx-maskrcnn (fetched 2026-08-28T04:05:31.830588+00:00, sha 1e9ae35210d9)
- Data as of 2026-08-30T08:39:29.467469+00:00.