Deep Unrolling of Sparsity-Induced RDO for 3D Point Cloud Attribute Coding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Do, Tam Thuc, Chou, Philip A., Cheung, Gene
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909780345356288
author Do, Tam Thuc
Chou, Philip A.
Cheung, Gene
author_facet Do, Tam Thuc
Chou, Philip A.
Cheung, Gene
contents Given encoded 3D point cloud geometry available at the decoder, we study the problem of lossy attribute compression in a multi-resolution B-spline projection framework. A target continuous 3D attribute function is first projected onto a sequence of nested subspaces $\mathcal{F}^{(p)}_{l_0} \subseteq \cdots \subseteq \mathcal{F}^{(p)}_{L}$, where $\mathcal{F}^{(p)}_{l}$ is a family of functions spanned by a B-spline basis function of order $p$ at a chosen scale and its integer shifts. The projected low-pass coefficients $F_l^*$ are computed by variable-complexity unrolling of a rate-distortion (RD) optimization algorithm into a feed-forward network, where the rate term is the sparsity-promoting $\ell_1$-norm. Thus, the projection operation is end-to-end differentiable. For a chosen coarse-to-fine predictor, the coefficients are then adjusted to account for the prediction from a lower-resolution to a higher-resolution, which is also optimized in a data-driven manner.
format Preprint
id arxiv_https___arxiv_org_abs_2509_08685
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Deep Unrolling of Sparsity-Induced RDO for 3D Point Cloud Attribute Coding
Do, Tam Thuc
Chou, Philip A.
Cheung, Gene
Image and Video Processing
Information Theory
Machine Learning
Given encoded 3D point cloud geometry available at the decoder, we study the problem of lossy attribute compression in a multi-resolution B-spline projection framework. A target continuous 3D attribute function is first projected onto a sequence of nested subspaces $\mathcal{F}^{(p)}_{l_0} \subseteq \cdots \subseteq \mathcal{F}^{(p)}_{L}$, where $\mathcal{F}^{(p)}_{l}$ is a family of functions spanned by a B-spline basis function of order $p$ at a chosen scale and its integer shifts. The projected low-pass coefficients $F_l^*$ are computed by variable-complexity unrolling of a rate-distortion (RD) optimization algorithm into a feed-forward network, where the rate term is the sparsity-promoting $\ell_1$-norm. Thus, the projection operation is end-to-end differentiable. For a chosen coarse-to-fine predictor, the coefficients are then adjusted to account for the prediction from a lower-resolution to a higher-resolution, which is also optimized in a data-driven manner.
title Deep Unrolling of Sparsity-Induced RDO for 3D Point Cloud Attribute Coding
topic Image and Video Processing
Information Theory
Machine Learning
url https://arxiv.org/abs/2509.08685