Ross ROSS = Recommend OSS · open-source software intelligence for agents

AI-FanGe/OpenAIglasses_for_Navigation

a open framework for blind navigation based on esp32 observed · 2026-08-28

github.com/AI-FanGe/OpenAIglasses_for_Navigation · Python · MIT (permissive) observed · 2026-08-28

Health v2 · maintenance only

39/100

  • Activity 50
  • Release rhythm 35
  • Longevity 22

Flags: no_releases

How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-02. Adoption (stars, forks) is never an input.

  • gap_med: n/a
  • age_days: 317
  • days_rel: n/a
  • days_push: 305
  • n_releases_24m: 0

Full methodology

Adoption not part of the score

1343 stars · 404 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

An open Python framework for an AI-powered smart glasses navigation system for visually impaired users, built around an ESP32-CAM client streaming video over WebSocket. It provides tactile paving (blind road) navigation, crosswalk and traffic light assistance, object finding with hand guidance, and real-time voice interaction using YOLO models, MediaPipe, and Alibaba DashScope/Qwen-Omni services.

Use cases

  • build assistive navigation glasses for blind users
  • detect tactile paving and guide a user along it with voice
  • help a visually impaired person cross streets by detecting crosswalks and traffic lights
  • find objects by voice command and guide the user's hand to them
  • stream ESP32 camera video to a web dashboard with real-time AI processing
  • fuse IMU data from ESP32 with vision for pose estimation
  • run real-time speech recognition and multimodal AI conversation on a live video feed

When to choose

  • you are prototyping or researching AI-based assistive navigation hardware
  • you want an integrated pipeline combining YOLO segmentation/detection, tracking, ASR, and multimodal LLM dialogue
  • you already have an ESP32-CAM or compatible WebSocket camera client and an NVIDIA GPU server
  • you want a Chinese-language voice interaction and monitoring interface

When to avoid

  • you need a production-ready, safety-certified device for real visually impaired users - the README explicitly warns against direct deployment
  • you have no NVIDIA GPU or cannot use Alibaba DashScope API keys
  • you need a lightweight embedded-only solution without a server-side Python stack
  • you need English or multilingual voice interfaces out of the box

Facets

framework · maturity active

computer-vision speech-recognition websocket machine-learning audio-processing chatbot computer-vision accessibility artificial-intelligence python cross-platform iot blind-navigation assistive-technology yolo esp32 tactile-paving-detection crosswalk-detection voice-guidance qwen-omni dashscope mediapipe bytetrack imu-fusion wearable-device real-time web-server gpu

1 source

Member repositories

RepositoryRoleHealth v2
AI-FanGe/OpenAIglasses_for_Navigationmain39

For agents

markdown · JSON · MCP: product_card(name="AI-FanGe/OpenAIglasses_for_Navigation")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem