# cmdbug/YOLOv5_NCNN

🍅 Deploy ncnn on mobile phones. Support Android and iOS. 移动端ncnn部署，支持Android与iOS。

Repository: https://github.com/cmdbug/YOLOv5_NCNN
Canonical: https://ross.abutalabs.com/products/yolov5_ncnn
Language: C++
License: GPL-3.0
License Family: copyleft
Topics: yolo, yolov5, ios, android, ncnn, detection, object-detection, yolov4, yolov3, mobilenet, simple-pose, yolact, chineseocr-lite, landmark106, dbface, mobilenetv3, nanodet, yolofastest, openpose
Last push: 2022-05-17T11:12:04+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": 2248, "days_push": 1569, "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 1572, forks 306 (observed 2026-08-28T04:05:05.510842+00:00)

## What it is
A mobile demo application that deploys the ncnn inference framework on Android and iOS, running a variety of computer vision models including YOLOv5s, YOLOv4-tiny, NanoDet, Yolact, ChineseOCR-lite, pose estimation, face detection, and segmentation models on the device camera. It serves as a reference implementation for converting and running PyTorch-trained models with ncnn on phones.

## Use cases
- run yolo object detection on android phone camera
- deploy ncnn models on ios
- test yolov5 on mobile device
- run pose estimation on smartphone
- on-device ocr with chineseocr-lite
- mobile semantic segmentation demo
- learn how to convert pytorch models to ncnn for mobile

## When to choose
- you want a working reference for deploying ncnn vision models on Android or iOS
- you need to benchmark lightweight detection, pose, OCR, or segmentation models on mobile CPUs/GPUs
- you are learning the ncnn mobile deployment workflow end to end

## When to avoid
- you need production-ready, actively maintained mobile inference code
- you target platforms other than Android or iOS
- you want server-side or desktop inference rather than on-device mobile deployment

## Facets
- artifact type: application
- maturity: maintenance
- function: computer-vision, image-processing, machine-learning, deep-learning
- domain: computer-vision, mobile-development, machine-learning, deep-learning
- platform: cpp
- tags: ncnn, yolov5, object-detection, on-device-inference, pose-estimation, ocr, semantic-segmentation, face-detection, demo-app, android, ios

## Member repositories
- cmdbug/YOLOv5_NCNN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:05.510842+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 2026-08-30T03:58:10.446503+00:00, confidence not recorded.
  - readme: https://github.com/cmdbug/YOLOv5_NCNN (fetched 2026-08-28T04:05:05.510842+00:00, sha 7eb4227a0474)
- Data as of 2026-08-30T08:39:29.467469+00:00.