Publication: Self-supervised monocular scene decomposition and depth estimation
Files
Program
KU-Authors
KU Authors
Co-Authors
Publication Date
Language
Embargo Status
NO
Journal Title
Journal ISSN
Volume Title
Alternative Title
Abstract
Self-supervised monocular depth estimation approaches either ignore independently moving objects in the scene or need a separate segmentation step to identify them. We propose MonoDepthSeg to jointly estimate depth and segment moving objects from monocular video without using any ground-truth labels. We decompose the scene into a fixed number of components where each component corresponds to a region on the image with its own transformation matrix representing its motion. We estimate both the mask and the motion of each component efficiently with a shared encoder. We evaluate our method on three driving datasets and show that our model clearly improves depth estimation while decomposing the scene into separately moving components.
Source
Publisher
IEEE Computer Society
Subject
Computer science, Engineering, Imaging science, Photographic technology
Citation
Has Part
Source
2021 International Conference on 3D Vision (3DV)
Book Series Title
Edition
DOI
10.1109/3DV53792.2021.00072