Salvato in:
Dettagli Bibliografici
Autori principali: Chidananda, Prajwal, Nair, Saurabh, Lee, Douglas, Kaehler, Adrian
Natura: Preprint
Pubblicazione: 2022
Soggetti:
Accesso online:https://arxiv.org/abs/2209.03910
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911777124515840
author Chidananda, Prajwal
Nair, Saurabh
Lee, Douglas
Kaehler, Adrian
author_facet Chidananda, Prajwal
Nair, Saurabh
Lee, Douglas
Kaehler, Adrian
contents We present PixTrack, a vision based object pose tracking framework using novel view synthesis and deep feature-metric alignment. We follow an SfM-based relocalization paradigm where we use a Neural Radiance Field to canonically represent the tracked object. Our evaluations demonstrate that our method produces highly accurate, robust, and jitter-free 6DoF pose estimates of objects in both monocular RGB images and RGB-D images without the need of any data annotation or trajectory smoothing. Our method is also computationally efficient making it easy to have multi-object tracking with no alteration to our algorithm through simple CPU multiprocessing. Our code is available at: https://github.com/GiantAI/pixtrack
format Preprint
id arxiv_https___arxiv_org_abs_2209_03910
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle PixTrack: Precise 6DoF Object Pose Tracking using NeRF Templates and Feature-metric Alignment
Chidananda, Prajwal
Nair, Saurabh
Lee, Douglas
Kaehler, Adrian
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Robotics
We present PixTrack, a vision based object pose tracking framework using novel view synthesis and deep feature-metric alignment. We follow an SfM-based relocalization paradigm where we use a Neural Radiance Field to canonically represent the tracked object. Our evaluations demonstrate that our method produces highly accurate, robust, and jitter-free 6DoF pose estimates of objects in both monocular RGB images and RGB-D images without the need of any data annotation or trajectory smoothing. Our method is also computationally efficient making it easy to have multi-object tracking with no alteration to our algorithm through simple CPU multiprocessing. Our code is available at: https://github.com/GiantAI/pixtrack
title PixTrack: Precise 6DoF Object Pose Tracking using NeRF Templates and Feature-metric Alignment
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Robotics
url https://arxiv.org/abs/2209.03910