Skip to content

occamLab/MiDaS

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

95 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Real-time object detection using LiDAR and image-based depth prediction

Overview

This repository contains an iPhone app that can be used for obstacle detection while navigating. The app currently works with LiDAR phones, and we are in the process of integrating the MiDaS neural network as another way to detect objects. MiDaS depth prediction is based on images captured by the camera instead of LiDAR. To see more of our work on MiDaS depth prediction and object detection, see the DepthBenchmarking repo. The app detects objects in the camera view and gives auditory feedback about the distance the closest object is from the camera. Currently, the user can toggle between feet and meters.

Usage

To use the app, clone the repository and follow the steps in the README in the ios folder to install all of the necessary packages. The app currently only works on iPhones equipped with LiDAR.

Acknowledgements

The base of our app was adopted from the original MiDaS repo and the work of this paper.

Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer
René Ranftl, Katrin Lasinger, David Hafner, Konrad Schindler, Vladlen Koltun

About

Implementation of MiDaS depth prediction on iPhone in conjunction with Apple's ARKit

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 9

Languages