Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Robust Full-Fov Depth Estimation in Tele-Wide Camera System

Abstract: 

Tele-wide camera system with different Field of View (FoV) lenses becomes very popular in recent mobile devices. Usually it is difficult to obtain full-FoV depth based on traditional stereo-matching methods. Pure Deep Neural Network (DNN) based depth estimation methods can obtain full-FoV depth, but have low robustness for scenarios which are not covered by training dataset. In this paper, to address the above problems we propose a hierarchical hourglass network for robust full-FoV depth estimation in tele-wide camera system, which combines the robustness of traditional stereo-matching methods with the accuracy of DNN. More specifically, the proposed network comprises three major modules: single image depth prediction module infers initial depth from input color image, depth propagation module propagates traditional stereo-matching tele-FoV depth to surrounding regions, and depth combination module fuses the initial depth with the propagated depth to generate final output. Each of these modules employs an hourglass model, which is a kind of encoder-decoder structure with skip connections. Experimental results compared with state-of-the-art depth estimation methods demonstrate that our method not only produces robust and better subjective depth quality on wild test images, but also obtains better quantitative results on standard datasets.

up
0 users have voted:

Paper Details

Authors:
Seongwook Song, Soonkeun Chang, Tae-ui Kim, Seungmin Han, Irina Kim
Submitted On:
14 May 2020 - 11:55am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Kai Guo
Paper Code:
1633
Document Year:
2020
Cite

Document Files

FoVdepth_presentation.pdf

(21)

Subscribe

[1] Seongwook Song, Soonkeun Chang, Tae-ui Kim, Seungmin Han, Irina Kim, "Robust Full-Fov Depth Estimation in Tele-Wide Camera System", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5310. Accessed: Jul. 13, 2020.
@article{5310-20,
url = {http://sigport.org/5310},
author = {Seongwook Song; Soonkeun Chang; Tae-ui Kim; Seungmin Han; Irina Kim },
publisher = {IEEE SigPort},
title = {Robust Full-Fov Depth Estimation in Tele-Wide Camera System},
year = {2020} }
TY - EJOUR
T1 - Robust Full-Fov Depth Estimation in Tele-Wide Camera System
AU - Seongwook Song; Soonkeun Chang; Tae-ui Kim; Seungmin Han; Irina Kim
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5310
ER -
Seongwook Song, Soonkeun Chang, Tae-ui Kim, Seungmin Han, Irina Kim. (2020). Robust Full-Fov Depth Estimation in Tele-Wide Camera System. IEEE SigPort. http://sigport.org/5310
Seongwook Song, Soonkeun Chang, Tae-ui Kim, Seungmin Han, Irina Kim, 2020. Robust Full-Fov Depth Estimation in Tele-Wide Camera System. Available at: http://sigport.org/5310.
Seongwook Song, Soonkeun Chang, Tae-ui Kim, Seungmin Han, Irina Kim. (2020). "Robust Full-Fov Depth Estimation in Tele-Wide Camera System." Web.
1. Seongwook Song, Soonkeun Chang, Tae-ui Kim, Seungmin Han, Irina Kim. Robust Full-Fov Depth Estimation in Tele-Wide Camera System [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5310