AVR: Augmented Vehicular Reality

AVR extends vehicular vision to see beyond obstruction by sharing views through V2V communications.

Note:

• We are actively working and updating this repo to incorporate more features and datasets.
• For those who want to play with the code, We will release a mini-dataset and release a stable version soon.

Getting started

Prerequisites

• Ubuntu 16.04
• Latest CMake
• ZED SDK 2.0
• CUDA 8.0
• sudo apt-get install libopenexr-dev libeigen3-dev libboost-all-dev libgtk2.0-dev libqt4-dev libopenni-dev libopenni2-dev libvtk6-dev libxmu-dev libxi-dev libflann-dev libcpprest-dev cmake libproj-dev
• PCL 1.7.2 installed with its visualization module
• Pangolin, get a release version
• ProtoBuf
• OpenCV 3.1 with OpenCV_Contrib, get a releaser version from both

Build the program

sh pre-build.sh  //build 3rd party libraries and untar vocabulary
mkdir build
cd build
cmake ..
make

Citation

@inproceedings{avr,
    author = {Qiu, Hang and Ahmad, Fawad and Bai, Fan and Gruteser, Marco and Govindan, Ramesh},
    title = {AVR: Augmented Vehicular Reality},
    year = {2018},
    url = {https://doi.org/10.1145/3210240.3210319},
    doi = {10.1145/3210240.3210319},
    booktitle = {Proceedings of the 16th Annual International Conference on Mobile Systems, Applications, and Services},
    pages = {81–95},
    numpages = {15},
    keywords = {Autonomous Cars, Extended Vision, Collaborative Sensing},
    location = {Munich, Germany},
    series = {MobiSys '18}
}