TY - GEN
T1 - Single-Image Depth Estimation Based on Fourier Domain Analysis
AU - Lee, Jae Han
AU - Heo, Minhyeok
AU - Kim, Kyung Rae
AU - Kim, Chang-Su
N1 - Funding Information:
This work was supported in part by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (NRF-2015R1A2A1A10055037, NRF-2018R1A2B3003896), and in part by the Agency for Defense Development (ADD) and Defense Acquisition Program Administration (DAPA) of Korea (UC160016FD).
Publisher Copyright:
© 2018 IEEE.
PY - 2018/12/14
Y1 - 2018/12/14
N2 - We propose a deep learning algorithm for single-image depth estimation based on the Fourier frequency domain analysis. First, we develop a convolutional neural network structure and propose a new loss function, called depth-balanced Euclidean loss, to train the network reliably for a wide range of depths. Then, we generate multiple depth map candidates by cropping input images with various cropping ratios. In general, a cropped image with a small ratio yields depth details more faithfully, while that with a large ratio provides the overall depth distribution more reliably. To take advantage of these complementary properties, we combine the multiple candidates in the frequency domain. Experimental results demonstrate that proposed algorithm provides the state-of-art performance. Furthermore, through the frequency domain analysis, we validate the efficacy of the proposed algorithm in most frequency bands.
AB - We propose a deep learning algorithm for single-image depth estimation based on the Fourier frequency domain analysis. First, we develop a convolutional neural network structure and propose a new loss function, called depth-balanced Euclidean loss, to train the network reliably for a wide range of depths. Then, we generate multiple depth map candidates by cropping input images with various cropping ratios. In general, a cropped image with a small ratio yields depth details more faithfully, while that with a large ratio provides the overall depth distribution more reliably. To take advantage of these complementary properties, we combine the multiple candidates in the frequency domain. Experimental results demonstrate that proposed algorithm provides the state-of-art performance. Furthermore, through the frequency domain analysis, we validate the efficacy of the proposed algorithm in most frequency bands.
UR - http://www.scopus.com/inward/record.url?scp=85055442364&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2018.00042
DO - 10.1109/CVPR.2018.00042
M3 - Conference contribution
AN - SCOPUS:85055442364
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 330
EP - 339
BT - Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018
PB - IEEE Computer Society
T2 - 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018
Y2 - 18 June 2018 through 22 June 2018
ER -