The KITTI Vision Benchmark Suite

we propose a multi-branch and high-level semantic
network by gradually splitting a base network into
multiple different branches.

see the paper:
https://ieeexplore.ieee.org/document/8887288

@inproceedings{jiale2018arXiv,
title={High-Level Semantic Networks for Multi-
Scale Object Detection},
author={Jiale Cao and Yanwei Pang and Shengjie
Zhao and Xuelong Li},
booktitle={IEEE Transactions on Circuits and
Systems for Video Technology},
year={2019}
}