2.5D Faster R-CNN is an extension of Faster R-CNN to enable it to detect the bounding-box of a target object and estimate its distance at the same time. To do this, 2D anchors in original Faster R-CNN are extended to 5 dimensional vectors including the distance candidate.
This repository provides codes for 2.5D Faster R-CNN proposed by Hirotaka Hachiya. The codes are based on py-faster-rcnn and written in python.
2.5D Faster R-CNN is released under the MIT License (refer to the LICENSE file for details).
If you find Faster R-CNN useful in your research, please consider citing:
@inproceedings{renNIPS15fasterrcnn,
Author = {Hirotaka Hachiya, Yuki Saito, Kazuma Iteya, Masaya Nomura and Takayuki Nakamura},
Title = {Distance estiamtion with 2.5D anchors and its application to robot navigation},
Booktitle = {arXiv},
Year = {2018}
}