TY - GEN
T1 - Monocular Depth Estimation Using Whole Strip Masking and Reliability-Based Refinement
AU - Heo, Minhyeok
AU - Lee, Jaehan
AU - Kim, Kyung Rae
AU - Kim, Han Ul
AU - Kim, Chang-Su
N1 - Funding Information:
This work was supported partly by the Cross-Ministry Giga KOREA Project Grant funded by the Korean Government (MSIT) (development of 4D reconstruction and dynamic deformable action model based hyper-realistic service technology) under Grant GK18P0200, partly by the National Research Foundation of Korea Grant funded by the Korean Government (MSIP) under Grant NRF-2015R1A2A1A10055037 and Grant NRF-2018R1A2B3003896, and partly by NAVER LABS.
Funding Information:
Acknowledgement. This work was supported partly by the Cross-Ministry Giga KOREA Project Grant funded by the Korean Government (MSIT) (development of 4D reconstruction and dynamic deformable action model based hyper-realistic service technology) under Grant GK18P0200, partly by the National Research Foundation of Korea Grant funded by the Korean Government (MSIP) under Grant NRF-2015R1A2A1A10055037 and Grant NRF-2018R1A2B3003896, and partly by NAVER LABS.
Publisher Copyright:
© 2018, Springer Nature Switzerland AG.
PY - 2018
Y1 - 2018
N2 - We propose a monocular depth estimation algorithm based on whole strip masking (WSM) and reliability-based refinement. First, we develop a convolutional neural network (CNN) tailored for the depth estimation. Specifically, we design a novel filter, called WSM, to exploit the tendency that a scene has similar depths in horizonal or vertical directions. The proposed CNN combines WSM upsampling blocks with a ResNet encoder. Second, we measure the reliability of an estimated depth, by appending additional layers to the main CNN. Using the reliability information, we perform conditional random field (CRF) optimization to refine the estimated depth map. Experimental results demonstrate that the proposed algorithm provides the state-of-the-art depth estimation performance.
AB - We propose a monocular depth estimation algorithm based on whole strip masking (WSM) and reliability-based refinement. First, we develop a convolutional neural network (CNN) tailored for the depth estimation. Specifically, we design a novel filter, called WSM, to exploit the tendency that a scene has similar depths in horizonal or vertical directions. The proposed CNN combines WSM upsampling blocks with a ResNet encoder. Second, we measure the reliability of an estimated depth, by appending additional layers to the main CNN. Using the reliability information, we perform conditional random field (CRF) optimization to refine the estimated depth map. Experimental results demonstrate that the proposed algorithm provides the state-of-the-art depth estimation performance.
KW - Depth map refinement
KW - Monocular depth estimation
KW - Reliability
KW - Whole strip masking
UR - http://www.scopus.com/inward/record.url?scp=85055438092&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-01225-0_3
DO - 10.1007/978-3-030-01225-0_3
M3 - Conference contribution
AN - SCOPUS:85055438092
SN - 9783030012243
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 39
EP - 55
BT - Computer Vision – ECCV 2018 - 15th European Conference, 2018, Proceedings
A2 - Ferrari, Vittorio
A2 - Sminchisescu, Cristian
A2 - Weiss, Yair
A2 - Hebert, Martial
PB - Springer Verlag
T2 - 15th European Conference on Computer Vision, ECCV 2018
Y2 - 8 September 2018 through 14 September 2018
ER -