Strivec: Sparse Tri-Vector Radiance Fields

TL;DR

Strivec introduces a sparse, local tensor-based neural radiance field representation using tri-vector factorization, improving rendering quality with fewer parameters by exploiting scene sparsity and multi-scale local features.

Contribution

The paper proposes a novel sparse local tensor representation with tri-vector factorization for neural radiance fields, enhancing efficiency and quality over prior global tensor methods.

Findings

Achieves better rendering quality than TensoRF and Instant-NGP.

Uses significantly fewer parameters due to scene sparsity.

Employs multi-scale local tensor grids for improved scene modeling.

Abstract

We propose Strivec, a novel neural representation that models a 3D scene as a radiance field with sparsely distributed and compactly factorized local tensor feature grids. Our approach leverages tensor decomposition, following the recent work TensoRF, to model the tensor grids. In contrast to TensoRF which uses a global tensor and focuses on their vector-matrix decomposition, we propose to utilize a cloud of local tensors and apply the classic CANDECOMP/PARAFAC (CP) decomposition to factorize each tensor into triple vectors that express local feature distributions along spatial axes and compactly encode a local neural field. We also apply multi-scale tensor grids to discover the geometry and appearance commonalities and exploit spatial coherence with the tri-vector factorization at multiple local scales. The final radiance field properties are regressed by aggregating neural features from multiple local tensors across all scales. Our tri-vector tensors are sparsely distributed around the actual scene surface, discovered by a fast coarse reconstruction, leveraging the sparsity of a 3D scene. We demonstrate that our model can achieve better rendering quality while using significantly fewer parameters than previous methods, including TensoRF and Instant-NGP.