RayFormer: Modeling Inter- and Intra-Ray Similarity for NeRF-Based Video Snapshot Compressive Imaging

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dong, Yubo, Liu, Danhua, Li, Anqi, Lin, Zhenyuan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918475577950208
author Dong, Yubo
Liu, Danhua
Li, Anqi
Lin, Zhenyuan
author_facet Dong, Yubo
Liu, Danhua
Li, Anqi
Lin, Zhenyuan
contents Video snapshot compressive imaging (SCI) enables the reconstruction of dynamic scenes from a single snapshot measurement. Recently, NeRF-based methods have shown promising reconstruction performance. However, such methods typically adopt random ray sampling strategies and fail to capture content structural similarities, resulting in limited reconstruction quality. To address these issues, we first propose a patch-level ray sampling strategy to enable the modeling of content structure. Then, we propose an Inter- and Intra-Ray Transformer (RayFormer) to capture the structural similarities, modeling both inter-ray similarities among spatially neighboring points at the same depth and intra-ray correlations between adjacent points along the viewing ray. Finally, benefiting from the patch-level sampling strategy, the total variation prior is incorporated into the objective function to enhance spatial smoothness and suppress artifacts. Experiments in both simulated and real-world scenes demonstrate that the proposed method achieves state-of-the-art (SOTA) reconstruction performance.
format Preprint
id arxiv_https___arxiv_org_abs_2604_27702
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RayFormer: Modeling Inter- and Intra-Ray Similarity for NeRF-Based Video Snapshot Compressive Imaging
Dong, Yubo
Liu, Danhua
Li, Anqi
Lin, Zhenyuan
Computer Vision and Pattern Recognition
Video snapshot compressive imaging (SCI) enables the reconstruction of dynamic scenes from a single snapshot measurement. Recently, NeRF-based methods have shown promising reconstruction performance. However, such methods typically adopt random ray sampling strategies and fail to capture content structural similarities, resulting in limited reconstruction quality. To address these issues, we first propose a patch-level ray sampling strategy to enable the modeling of content structure. Then, we propose an Inter- and Intra-Ray Transformer (RayFormer) to capture the structural similarities, modeling both inter-ray similarities among spatially neighboring points at the same depth and intra-ray correlations between adjacent points along the viewing ray. Finally, benefiting from the patch-level sampling strategy, the total variation prior is incorporated into the objective function to enhance spatial smoothness and suppress artifacts. Experiments in both simulated and real-world scenes demonstrate that the proposed method achieves state-of-the-art (SOTA) reconstruction performance.
title RayFormer: Modeling Inter- and Intra-Ray Similarity for NeRF-Based Video Snapshot Compressive Imaging
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.27702