Home

Awesome

PWC

EgoNet

Official project website for the CVPR 2021 paper "Exploring intermediate representation for monocular vehicle pose estimation". This repo includes an implementation that performs vehicle orientation estimation on the KITTI dataset from a single RGB image.

News:

(2022-??-??): v-1.1 will be released which include pre-trained models for other object classes (Pedestrian and Cyclist in KITTI).

(2021-08-16): v-1.0 is released. The training documentation is added.

(2021-06-21): v-0.9 (beta version) is released. The inference utility is here! For Q&A, go to discussions. If you believe there is a technical problem, submit to issues.

(2021-06-16): This repo is under final code cleaning and documentation preparation. Stay tuned and come back in a week!

Check our 5-min video (Youtube, 爱奇艺) for an introduction.

中文详解哔哩哔哩

<p align="center"> <img src="https://github.com/Nicholasli1995/EgoNet/blob/master/imgs/teaser.jpg" width="830" height="200" /> </p>

Run a demo with a one-line command!

Check instructions here.

<p align="center"> <img src="https://github.com/Nicholasli1995/EgoNet/blob/master/imgs/Ego-Net_demo.png" height="175"/> <img src="https://github.com/Nicholasli1995/EgoNet/blob/master/imgs/Ego-Net_demo.gif" height="175"/> </p>

Performance: AP<sup>BEV</sup>@R<sub>40</sub> on KITTI val set for Car (monocular RGB)

The validation results in the paper was based on R<sub>11</sub>, the results using R<sub>40</sub> are attached here.

MethodReferenceEasyModerateHard
M3D-RPNICCV 201920.8515.6211.88
MonoDISICCV 201918.4512.5810.66
MonoPairCVPR 202024.1218.1715.76
D4LCNCVPR 202031.5322.5817.87
Kinematic3DECCV 202027.8319.7215.10
GrooMeD-NMSCVPR 202127.3819.7515.92
MonoDLECVPR 202124.9719.3317.01
Ours (@R<sub>11</sub>)CVPR 202133.6025.3822.80
Ours (@R<sub>40</sub>)CVPR 202134.3124.8020.16

Performance: AOS@R<sub>40</sub> on KITTI test set for Car (RGB)

MethodReferenceConfigurationEasyModerateHard
M3D-RPNICCV 2019Monocular88.3882.8167.08
DSGNCVPR 2020Stereo95.4286.0378.27
Disp-RCNNCVPR 2020Stereo93.0281.7067.16
MonoPairCVPR 2020Monocular91.6586.1176.45
D4LCNCVPR 2020Monocular90.0182.0863.98
Kinematic3DECCV 2020Monocular58.3345.5034.81
MonoDLECVPR 2021Monocular93.4690.2380.11
OursCVPR 2021Monocular96.1191.2380.96

Inference/Deployment

Check instructions here to reproduce the above quantitative results.

Training

Check instructions here to train Ego-Net and learn how to prepare your own training dataset other than KITTI.

Citation

Please star this repository and cite the following paper in your publications if it helps your research:

@InProceedings{Li_2021_CVPR,
author    = {Li, Shichao and Yan, Zengqiang and Li, Hongyang and Cheng, Kwang-Ting},
title     = {Exploring intermediate representation for monocular vehicle pose estimation},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month     = {June},
year      = {2021},
pages     = {1873-1883}
}

License

This repository can be used freely for non-commercial purposes. Contact me if you are interested in a commercial license.

Links

Link to the paper: Exploring intermediate representation for monocular vehicle pose estimation

Link to the presentation video: Youtube, 爱奇艺

Relevant ECCV 2020 work: GSNet