Mobileye - Mobile eye for visually impared

Mobileye is 3rd winner in {M1522.006600}Intelligent System Design Project, CSE, SNU.

This demo application works mono depth estimation in mobile phone with fast-depth.

It gives vibration notification to user when some hazards like passing cars, motorcycles, even some man with close walk in their real life occurs.

Demo Video:

Backgrounds

We implement this repo with some backgrounds in PyTorchDemoApp.

From main view, with Vision Processing example in main view, you can start some fast-depth options with full convolution or depth-wise separable convolution.

Fast Start

Install app-debug.apk file with link

Environment

Java17.0.1: Download

Android10: Set android studio sdk manager as android10.(Tools - SDK manager)

pytorch_android:1.8.0

pytorch_android_torchvision:1.8.0

Settings

(Gradle version) In ~/gradle.wrapper/gradle-wrapper.properties

distributionUrl=https://services.gradle.org/distributions/gradle-7.2-all.zip

(Gradle plugin) In ~/build.gradle

dependencies{
	classpath'com.android.tools.build:gradle:7.0.0'
}

(Pytorch verision) In app/build.gradle

dependencies {
    implementation 'org.pytorch:pytorch_android:1.8.0'
    implementation 'org.pytorch:pytorch_android_torchvision:1.8.0'
}

How to build

Clone or download this repository:

git clone https://github.com/HanByulKim/Mobileye

Clean gradle build:

./gradlew clean

Build gradle with output apk file (path: ~/app/build/outputs/apk/debug/app-debug.apk):

./gradlew assembleDebug

Citation

If you find our project helpful, please consider to cite our project

@article{Mobileye2021,
  title = {{{Mobileye}}: {{Visualizing DNN Quantization effect on Network.}},
  shorttitle = {{{Mobileye}}},
  author = {Han-Byul Kim and Seunghun Shin},
  project={2021SNU_ISD},
  year={2021}
}