Method

Continuous Depth Network [st] [CDN]
https://github.com/Div99/W-Stereo-Disp

Submitted on 2 Jun. 2020 19:35 by
Naman Garg (Binghamton)

Running time:0.6 s
Environment:GPU @ 2.5 Ghz (Python)

Method Description:
Wasserstein Distances for Stereo Disparity
Estimation
Parameters:
with DSGN
Latex Bibtex:
@InProceedings{garg2020wasserstein,
title={Wasserstein Distances for Stereo
Disparity Estimation},
author={Divyansh Garg and Yan Wang and
Bharath Hariharan and Mark Campbell and Kilian Q.
Weinberger and Wei-Lun Chao},
booktitle = {Advances in Neural
Information Processing Systems (NeurIPS)},
year = {2020}
}

Detailed Results

Object detection and orientation estimation results. Results for object detection are given in terms of average precision (AP) and results for joint object detection and orientation estimation are provided in terms of average orientation similarity (AOS).


Benchmark Easy Moderate Hard
Car (Detection) 95.85 % 87.19 % 79.43 %
Car (Orientation) 95.79 % 86.90 % 79.05 %
Car (3D Detection) 74.52 % 54.22 % 46.36 %
Car (Bird's Eye View) 83.32 % 66.24 % 57.65 %
This table as LaTeX


2D object detection results.
This figure as: png eps pdf txt gnuplot



Orientation estimation results.
This figure as: png eps pdf txt gnuplot



3D object detection results.
This figure as: png eps pdf txt gnuplot



Bird's eye view results.
This figure as: png eps pdf txt gnuplot




eXTReMe Tracker