Saved in:
Bibliographic Details
Main Authors: Mazur, Kirill, Bae, Gwangbin, Davison, Andrew J.
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.05889
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909172235239424
author Mazur, Kirill
Bae, Gwangbin
Davison, Andrew J.
author_facet Mazur, Kirill
Bae, Gwangbin
Davison, Andrew J.
contents Joint camera pose and dense geometry estimation from a set of images or a monocular video remains a challenging problem due to its computational complexity and inherent visual ambiguities. Most dense incremental reconstruction systems operate directly on image pixels and solve for their 3D positions using multi-view geometry cues. Such pixel-level approaches suffer from ambiguities or violations of multi-view consistency (e.g. caused by textureless or specular surfaces). We address this issue with a new image representation which we call a SuperPrimitive. SuperPrimitives are obtained by splitting images into semantically correlated local regions and enhancing them with estimated surface normal directions, both of which are predicted by state-of-the-art single image neural networks. This provides a local geometry estimate per SuperPrimitive, while their relative positions are adjusted based on multi-view observations. We demonstrate the versatility of our new representation by addressing three 3D reconstruction tasks: depth completion, few-view structure from motion, and monocular dense visual odometry.
format Preprint
id arxiv_https___arxiv_org_abs_2312_05889
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle SuperPrimitive: Scene Reconstruction at a Primitive Level
Mazur, Kirill
Bae, Gwangbin
Davison, Andrew J.
Computer Vision and Pattern Recognition
Joint camera pose and dense geometry estimation from a set of images or a monocular video remains a challenging problem due to its computational complexity and inherent visual ambiguities. Most dense incremental reconstruction systems operate directly on image pixels and solve for their 3D positions using multi-view geometry cues. Such pixel-level approaches suffer from ambiguities or violations of multi-view consistency (e.g. caused by textureless or specular surfaces). We address this issue with a new image representation which we call a SuperPrimitive. SuperPrimitives are obtained by splitting images into semantically correlated local regions and enhancing them with estimated surface normal directions, both of which are predicted by state-of-the-art single image neural networks. This provides a local geometry estimate per SuperPrimitive, while their relative positions are adjusted based on multi-view observations. We demonstrate the versatility of our new representation by addressing three 3D reconstruction tasks: depth completion, few-view structure from motion, and monocular dense visual odometry.
title SuperPrimitive: Scene Reconstruction at a Primitive Level
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.05889