Dual-branch Graph Feature Learning for NLOS Imaging

TL;DR

This paper presents xnet, a dual-branch GNN-based framework for NLOS imaging that improves reconstruction quality and efficiency by separately processing albedo and depth information, outperforming existing methods.

Contribution

The paper introduces xnet, the first to use GNNs for transforming dense NLOS grid data into sparse features, enabling efficient and high-quality reconstruction of occluded scenes.

Findings

Achieves state-of-the-art performance on synthetic and real datasets.

Effectively separates albedo and depth reconstruction for improved results.

Reduces computational and storage requirements compared to prior methods.

Abstract

The domain of non-line-of-sight (NLOS) imaging is advancing rapidly, offering the capability to reveal occluded scenes that are not directly visible. However, contemporary NLOS systems face several significant challenges: (1) The computational and storage requirements are profound due to the inherent three-dimensional grid data structure, which restricts practical application. (2) The simultaneous reconstruction of albedo and depth information requires a delicate balance using hyperparameters in the loss function, rendering the concurrent reconstruction of texture and depth information difficult. This paper introduces the innovative methodology, \xnet, which integrates an albedo-focused reconstruction branch dedicated to albedo information recovery and a depth-focused reconstruction branch that extracts geometrical structure, to overcome these obstacles. The dual-branch framework segregates content delivery to the respective reconstructions, thereby enhancing the quality of the retrieved data. To our knowledge, we are the first to employ the GNN as a fundamental component to transform dense NLOS grid data into sparse structural features for efficient reconstruction. Comprehensive experiments demonstrate that our method attains the highest level of performance among existing methods across synthetic and real data. https://github.com/Nicholassu/DG-NLOS.