Neural Light Spheres for Implicit Image Stitching and View Synthesis

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chugunov, Ilya, Joshi, Amogh, Murthy, Kiran, Bleibel, Francois, Heide, Felix
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909554274467840
author Chugunov, Ilya
Joshi, Amogh
Murthy, Kiran
Bleibel, Francois
Heide, Felix
author_facet Chugunov, Ilya
Joshi, Amogh
Murthy, Kiran
Bleibel, Francois
Heide, Felix
contents Challenging to capture, and challenging to display on a cellphone screen, the panorama paradoxically remains both a staple and underused feature of modern mobile camera applications. In this work we address both of these challenges with a spherical neural light field model for implicit panoramic image stitching and re-rendering; able to accommodate for depth parallax, view-dependent lighting, and local scene motion and color changes during capture. Fit during test-time to an arbitrary path panoramic video capture -- vertical, horizontal, random-walk -- these neural light spheres jointly estimate the camera path and a high-resolution scene reconstruction to produce novel wide field-of-view projections of the environment. Our single-layer model avoids expensive volumetric sampling, and decomposes the scene into compact view-dependent ray offset and color components, with a total model size of 80 MB per scene, and real-time (50 FPS) rendering at 1080p resolution. We demonstrate improved reconstruction quality over traditional image stitching and radiance field methods, with significantly higher tolerance to scene motion and non-ideal capture settings.
format Preprint
id arxiv_https___arxiv_org_abs_2409_17924
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Neural Light Spheres for Implicit Image Stitching and View Synthesis
Chugunov, Ilya
Joshi, Amogh
Murthy, Kiran
Bleibel, Francois
Heide, Felix
Computer Vision and Pattern Recognition
Challenging to capture, and challenging to display on a cellphone screen, the panorama paradoxically remains both a staple and underused feature of modern mobile camera applications. In this work we address both of these challenges with a spherical neural light field model for implicit panoramic image stitching and re-rendering; able to accommodate for depth parallax, view-dependent lighting, and local scene motion and color changes during capture. Fit during test-time to an arbitrary path panoramic video capture -- vertical, horizontal, random-walk -- these neural light spheres jointly estimate the camera path and a high-resolution scene reconstruction to produce novel wide field-of-view projections of the environment. Our single-layer model avoids expensive volumetric sampling, and decomposes the scene into compact view-dependent ray offset and color components, with a total model size of 80 MB per scene, and real-time (50 FPS) rendering at 1080p resolution. We demonstrate improved reconstruction quality over traditional image stitching and radiance field methods, with significantly higher tolerance to scene motion and non-ideal capture settings.
title Neural Light Spheres for Implicit Image Stitching and View Synthesis
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.17924