TY - GEN
T1 - Deep stereo confidence prediction for depth estimation
AU - Kim, Sunok
AU - Min, Dongbo
AU - Ham, Bumsub
AU - Kim, Seungryong
AU - Sohn, Kwanghoon
N1 - Funding Information:
This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIP) (NRF-2016R1A2A2A05921659).
Publisher Copyright:
© 2017 IEEE.
PY - 2018/2/20
Y1 - 2018/2/20
N2 - We present a novel method that predicts a confidence to improve the accuracy of an estimated depth map in stereo matching. In contrast to existing learning based approaches relying on hand-crafted confidence features, we cast this problem into a convolutional neural network, learned using both a matching cost volume and its associated disparity map. As the size of the matching cost volume varies depending on a search range of stereo image pairs, we propose to use a top-K matching probability volume layer so that an input size for convolutional layers remains unchanged. Experimental results demonstrate that the proposed method outperforms the state-of-the-art confidence estimation approaches on various benchmarks.
AB - We present a novel method that predicts a confidence to improve the accuracy of an estimated depth map in stereo matching. In contrast to existing learning based approaches relying on hand-crafted confidence features, we cast this problem into a convolutional neural network, learned using both a matching cost volume and its associated disparity map. As the size of the matching cost volume varies depending on a search range of stereo image pairs, we propose to use a top-K matching probability volume layer so that an input size for convolutional layers remains unchanged. Experimental results demonstrate that the proposed method outperforms the state-of-the-art confidence estimation approaches on various benchmarks.
KW - Confidence prediction
KW - Convolutional neural networks
KW - Depth refinement
KW - Matching probability
KW - Stereo matching
UR - http://www.scopus.com/inward/record.url?scp=85045311133&partnerID=8YFLogxK
U2 - 10.1109/ICIP.2017.8296430
DO - 10.1109/ICIP.2017.8296430
M3 - Conference contribution
AN - SCOPUS:85045311133
T3 - Proceedings - International Conference on Image Processing, ICIP
SP - 992
EP - 996
BT - 2017 IEEE International Conference on Image Processing, ICIP 2017 - Proceedings
PB - IEEE Computer Society
Y2 - 17 September 2017 through 20 September 2017
ER -