Deep Non-rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Deng, Hui, Shi, Jiawei, Qin, Zhen, Zhong, Yiran, Dai, Yuchao
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915056091922432
author Deng, Hui
Shi, Jiawei
Qin, Zhen
Zhong, Yiran
Dai, Yuchao
author_facet Deng, Hui
Shi, Jiawei
Qin, Zhen
Zhong, Yiran
Dai, Yuchao
contents Non-Rigid Structure-from-Motion (NRSfM) is a classic 3D vision problem, where a 2D sequence is taken as input to estimate the corresponding 3D sequence. Recently, the deep neural networks have greatly advanced the task of NRSfM. However, existing deep NRSfM methods still have limitations in handling the inherent sequence property and motion ambiguity associated with the NRSfM problem. In this paper, we revisit deep NRSfM from two perspectives to address the limitations of current deep NRSfM methods : (1) canonicalization and (2) sequence modeling. We propose an easy-to-implement per-sequence canonicalization method as opposed to the previous per-dataset canonicalization approaches. With this in mind, we propose a sequence modeling method that combines temporal information and subspace constraint. As a result, we have achieved a more optimal NRSfM reconstruction pipeline compared to previous efforts. The effectiveness of our method is verified by testing the sequence-to-sequence deep NRSfM pipeline with corresponding regularization modules on several commonly used datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2412_07230
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Deep Non-rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling
Deng, Hui
Shi, Jiawei
Qin, Zhen
Zhong, Yiran
Dai, Yuchao
Computer Vision and Pattern Recognition
Non-Rigid Structure-from-Motion (NRSfM) is a classic 3D vision problem, where a 2D sequence is taken as input to estimate the corresponding 3D sequence. Recently, the deep neural networks have greatly advanced the task of NRSfM. However, existing deep NRSfM methods still have limitations in handling the inherent sequence property and motion ambiguity associated with the NRSfM problem. In this paper, we revisit deep NRSfM from two perspectives to address the limitations of current deep NRSfM methods : (1) canonicalization and (2) sequence modeling. We propose an easy-to-implement per-sequence canonicalization method as opposed to the previous per-dataset canonicalization approaches. With this in mind, we propose a sequence modeling method that combines temporal information and subspace constraint. As a result, we have achieved a more optimal NRSfM reconstruction pipeline compared to previous efforts. The effectiveness of our method is verified by testing the sequence-to-sequence deep NRSfM pipeline with corresponding regularization modules on several commonly used datasets.
title Deep Non-rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.07230