# WisconsinAIVision/yolact_edge

The first competitive instance segmentation approach that runs on small edge devices at real-time speeds.

Repository: https://github.com/WisconsinAIVision/yolact_edge
Canonical: https://ross.abutalabs.com/products/yolact_edge
Language: Python
License: MIT
License Family: permissive
Topics: realtime, real-time, instance-segmentation, yolactedge, pytorch, edge-devices
Last push: 2022-12-07T02:53:17+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2080, "days_push": 1365, "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 1322, forks 274 (observed 2026-08-28T04:04:22.056475+00:00)

## What it is
YolactEdge is a PyTorch implementation of a real-time instance segmentation model optimized for edge devices like the NVIDIA Jetson AGX Xavier. It provides pretrained weights, evaluation, and training code for COCO and YouTube VIS benchmarks.

## Use cases
- run real-time instance segmentation on edge devices
- segment objects in video at high FPS on a Jetson
- evaluate pretrained instance segmentation models on COCO
- train a fast instance segmentation model with a ResNet backbone
- deploy computer vision on resource-constrained hardware

## When to choose
- you need real-time instance segmentation on embedded GPUs like Jetson
- you want a PyTorch codebase with pretrained COCO/YouTube VIS weights
- you can leverage TensorRT for maximum edge inference speed

## When to avoid
- you need a maintained library with frequent updates (last release 2022)
- you need panoptic or semantic segmentation rather than instance segmentation
- you only target CPU-only or non-NVIDIA hardware

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning
- platform: python, embedded
- tags: instance-segmentation, real-time-inference, edge-devices, pytorch, tensorrt, yolact, jetson, linux, gpu

## Member repositories
- WisconsinAIVision/yolact_edge (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:22.056475+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 2026-08-30T04:47:06.960957+00:00, confidence not recorded.
  - readme: https://github.com/WisconsinAIVision/yolact_edge (fetched 2026-08-28T04:04:22.056475+00:00, sha 7380b5d1e90d)
- Data as of 2026-08-30T08:39:29.467469+00:00.