Epipolar Geometry based Learning of Multi-view Depth and Ego-Motion from Monocular Sequences

12/23/2018
by   Vignesh Prasad, et al.
0

Deep approaches to predict monocular depth and ego-motion have grown in recent years due to their ability to produce dense depth from monocular images. The main idea behind them is to optimize the photometric consistency over image sequences by warping one view into another, similar to direct visual odometry methods. One major drawback is that these methods infer depth from a single view, which might not effectively capture the relation between pixels. Moreover, simply minimizing the photometric loss does not ensure proper pixel correspondences, which is a key factor for accurate depth and pose estimations. In contrast, we propose a 2-view depth network to infer the scene depth from consecutive frames, thereby learning inter-pixel relationships. To ensure better correspondences, thereby better geometric understanding, we propose incorporating epipolar constraints to make the learning more geometrically sound. We use the Essential matrix obtained using Nist'er's Five Point Algorithm, to enforce meaningful geometric constraints, rather than using it as training labels. This allows us to use lesser no. of trainable parameters compared to state-of-the-art methods. The proposed method results in better depth images and pose estimates, which capture the scene structure and motion in a better way. Such a geometrically constrained learning performs successfully even in cases where simply minimizing the photometric error would fail.

READ FULL TEXT

page 3

page 5

page 8

page 10

research
12/20/2018

SfMLearner++: Learning Monocular Depth & Ego-Motion using Meaningful Geometric Constraints

Most geometric approaches to monocular Visual Odometry (VO) provide robu...
research
08/17/2019

Mono-SF: Multi-View Geometry Meets Single-View Depth for Monocular Scene Flow Estimation of Dynamic Traffic Scenes

Existing 3D scene flow estimation methods provide the 3D geometry and 3D...
research
08/05/2016

Photometric Bundle Adjustment for Vision-Based SLAM

We propose a novel algorithm for the joint refinement of structure and m...
research
12/06/2021

3D Hierarchical Refinement and Augmentation for Unsupervised Learning of Depth and Pose from Monocular Video

Depth and ego-motion estimations are essential for the localization and ...
research
08/04/2021

Self-Supervised Learning of Depth and Ego-Motion from Video by Alternative Training and Geometric Constraints from 3D to 2D

Self-supervised learning of depth and ego-motion from unlabeled monocula...
research
03/14/2022

RAUM-VO: Rotational Adjusted Unsupervised Monocular Visual Odometry

Unsupervised learning for monocular camera motion and 3D scene understan...
research
12/01/2017

Learning Depth from Monocular Videos using Direct Methods

The ability to predict depth from a single image - using recent advances...

Please sign up or login with your details

Forgot password? Click here to reset