UNav is designed for helping navigation of visually impaired people
Project description
:rocket: UNav
UNav is a vision-based location system designed to assist visually impaired individuals in navigating unfamiliar environments.
:sparkles: New Features
- May 29, 2023. Support Parallel RanSAC computing
More
:wrench: Dependencies and Installation
- Python >= 3.8 (Recommend to use Anaconda or Miniconda)
- PyTorch >= 1.13
- NVIDIA GPU + CUDA
-
Clone repo
git clone https://github.com/endeleze/UNav.git
-
Install dependent packages
cd UNav pip install -r requirements.txt
:computer: Using
-
Server-Client
-
Setup server.yaml and tune hloc.yaml us needed.
-
Put the data into IO_root you defined as following structure
UNav-IO/ ├── data │ ├── destination.json │ ├── PLACE │ │ └── BUILDING │ │ └── FLOOR │ │ ├── access_graph.npy │ │ ├── boundaries.json │ │ ├── feats-superpoint.h5 │ │ ├── global_features.h5 │ │ ├── topo-map.json │ │ └── floorplan.png
Note that you need to rerun Path_finder_waypoints.py using step2_automatically.sh if you do not have access_graph.npy
-
Run server using
source shell/server.sh
-
Run client device
- Jetson Board
- Android
-
-
Visualization-GUI TODO
Note that UNav is only tested in Ubuntu, and may be not suitable for Windows or MacOS.
:hourglass_flowing_sand: TODO List
Please see project boards.
:e-mail: Contact
If you have any question, please email ay1620@nyu.edu
.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.