CVApr 3, 2022

BNV-Fusion: Dense 3D Reconstruction using Bi-level Neural Volume Fusion

arXiv:2204.01139v152 citationsh-index: 117Has Code
Originality Incremental advance
AI Analysis

This addresses the problem of balancing robustness and detail in 3D reconstruction for mixed reality and robotic applications, representing an incremental advancement.

The paper tackled dense 3D reconstruction from depth images by proposing BNV-Fusion, a method using a bi-level neural volume fusion strategy, which demonstrated significant improvement over existing methods in evaluations on multiple datasets.

Dense 3D reconstruction from a stream of depth images is the key to many mixed reality and robotic applications. Although methods based on Truncated Signed Distance Function (TSDF) Fusion have advanced the field over the years, the TSDF volume representation is confronted with striking a balance between the robustness to noisy measurements and maintaining the level of detail. We present Bi-level Neural Volume Fusion (BNV-Fusion), which leverages recent advances in neural implicit representations and neural rendering for dense 3D reconstruction. In order to incrementally integrate new depth maps into a global neural implicit representation, we propose a novel bi-level fusion strategy that considers both efficiency and reconstruction quality by design. We evaluate the proposed method on multiple datasets quantitatively and qualitatively, demonstrating a significant improvement over existing methods.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes