TY - GEN
T1 - Road perspective depth reconstruction from single images using reduce-refine-upsample CNNs
AU - Valdez-Rodríguez, José E.
AU - Calvo, Hiram
AU - Felipe-Riverón, Edgardo M.
N1 - Publisher Copyright:
© Springer Nature Switzerland AG 2018.
PY - 2018
Y1 - 2018
N2 - Depth reconstruction from single images has been a challenging task due to the complexity and the quantity of depth cues that images have. Convolutional Neural Networks (CNN) have been successfully used to reconstruct depth of general object scenes; however, these works have not been tailored for the particular problem of road perspective depth reconstruction. As we aim to build a computational efficient model, we focus on single-stage CNNs. In this paper we propose two different models for solving this task. A particularity is that our models perform refinement in the same single-stage training; thus, we call them Reduce-Refine-Upsample (RRU) models because of the order of the CNN operations. We compare our models with the current state of the art in depth reconstruction, obtaining improvements in both global and local views for images of road perspectives.
AB - Depth reconstruction from single images has been a challenging task due to the complexity and the quantity of depth cues that images have. Convolutional Neural Networks (CNN) have been successfully used to reconstruct depth of general object scenes; however, these works have not been tailored for the particular problem of road perspective depth reconstruction. As we aim to build a computational efficient model, we focus on single-stage CNNs. In this paper we propose two different models for solving this task. A particularity is that our models perform refinement in the same single-stage training; thus, we call them Reduce-Refine-Upsample (RRU) models because of the order of the CNN operations. We compare our models with the current state of the art in depth reconstruction, obtaining improvements in both global and local views for images of road perspectives.
KW - Convolutional neural networks
KW - Depth reconstruction
KW - Embedded refining layer
KW - One stage training
KW - Stereo matching
UR - http://www.scopus.com/inward/record.url?scp=85059958332&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-02837-4_3
DO - 10.1007/978-3-030-02837-4_3
M3 - Contribución a la conferencia
AN - SCOPUS:85059958332
SN - 9783030028367
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 30
EP - 40
BT - Advances in Soft Computing - 16th Mexican International Conference on Artificial Intelligence, MICAI 2017, Proceedings
A2 - Miranda-Jiménez, Sabino
A2 - Castro, Félix
A2 - González-Mendoza, Miguel
PB - Springer Verlag
T2 - 16th Mexican International Conference on Artificial Intelligence, MICAI 2017
Y2 - 23 October 2017 through 28 October 2017
ER -