On Regression Losses for Deep Depth Estimation

Depth estimation from a single monocular image has reached great performances thanks to recent works based on deep networks. However, as various choices of losses, architectures and experimental conditions are proposed in the literature, it is difficult to establish their respective influence on the...

Full description

Saved in:
Bibliographic Details
Main Authors: Carvalho, Marcela, Saux, Bertrand Le, Trouve-Peloux, Pauline, Almansa, Andres, Champagnat, Frederic
Format: Conference Proceeding
Language:English
Subjects:
Convolution
deep learning
Depth estimation
Estimation
Gallium nitride
loss function
Machine learning
Network architecture
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Depth estimation from a single monocular image has reached great performances thanks to recent works based on deep networks. However, as various choices of losses, architectures and experimental conditions are proposed in the literature, it is difficult to establish their respective influence on the performances. In this paper we propose an in-depth study of various losses and experimental conditions for depth regression, on NYUv2 dataset. From this study we propose a new network for depth estimation combining an encoder-decoder architecture with an adversarial loss. This network reaches top scores in the competitive evaluation of NUYv2 dataset while being simpler to train in a single phase.
ISSN:2381-8549
DOI:10.1109/ICIP.2018.8451312