Detail preserving depth prediction from a single image based on multi-scale deep network and gradient network

Huihui Xu; Liu Nan

doi:10.1117/12.2623204

16 February 2022 Detail preserving depth prediction from a single image based on multi-scale deep network and gradient network

Huihui Xu, Liu Nan

Author Affiliations +

Proceedings Volume 12083, Thirteenth International Conference on Graphics and Image Processing (ICGIP 2021); 120830J (2022) https://doi.org/10.1117/12.2623204
Event: Thirteenth International Conference on Graphics and Image Processing (ICGIP 2021), 2021, Kunming, China

Abstract

redicting a convincing depth map from a monocular RGB image is a daunting task in the field of computer vision. An efficient detail preserving depth prediction algorithm on the basis of multi-scale deep network and gradient network is presented in this paper. Specifically, the proposed method leverage the designed multi-scale deep network to obtain the global depth image from training datasets. Moreover, a depth gradient generation strategy on the basis of gradient network, which permits us to obtain the local depth detail information of the scene is developed. In the end, the reliable depth map with better details could be reconstructed via merging the depth gradient and depth information on the basis of the optimization algorithm. Experimental results evaluated from the qualitative and quantitative perspective on the Make 3D and NYUv2 datasets indicate that the designed depth prediction scheme is superior to several depth estimation approaches, and can reconstruct plausible depths

Citation Download Citation

Huihui Xu and Liu Nan "Detail preserving depth prediction from a single image based on multi-scale deep network and gradient network", Proc. SPIE 12083, Thirteenth International Conference on Graphics and Image Processing (ICGIP 2021), 120830J (16 February 2022); https://doi.org/10.1117/12.2623204

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
10 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Convolution

Network architectures

3D image processing

Feature extraction

Machine vision

RELATED CONTENT

An optimized feature extraction algorithm based on SuperPoint
Proceedings of SPIE (January 20 2023)

A 3D reconstruction algorithm based on adaptive cost aggregation and...
Proceedings of SPIE (October 20 2022)

Real-time stereo matching model based on deformable convolution
Proceedings of SPIE (October 22 2021)

Real-time traffic sign detection based on YOLOv2
Proceedings of SPIE (October 29 2018)

Shrimps classification based on multi-layer feature fusion
Proceedings of SPIE (May 06 2019)

Blood vessels segmentation of hatching eggs based on fully convolutional...
Proceedings of SPIE (April 10 2018)

Hybrid connection network for semantic segmentation
Proceedings of SPIE (August 09 2018)

Subscribe to Digital Library

Receive Erratum Email Alert

Keywords/Phrases

Search In:

Publication Years