Tensorformer: Normalized Matrix Attention Transformer for High-quality Point Cloud Reconstruction

TL;DR

Tensorformer introduces a normalized matrix attention transformer that enhances local geometry modeling for high-quality point cloud surface reconstruction, outperforming previous methods on standard datasets.

Contribution

The paper proposes a novel normalized matrix attention mechanism within a transformer architecture, enabling simultaneous point-wise and channel-wise message passing for improved reconstruction.

Findings

Achieves state-of-the-art results on ShapeNetCore and ABC datasets.

Attains 4% improvement in IOU on ShapeNet.

Outperforms existing methods in surface reconstruction quality.

Abstract

Surface reconstruction from raw point clouds has been studied for decades in the computer graphics community, which is highly demanded by modeling and rendering applications nowadays. Classic solutions, such as Poisson surface reconstruction, require point normals as extra input to perform reasonable results. Modern transformer-based methods can work without normals, while the results are less fine-grained due to limited encoding performance in local fusion from discrete points. We introduce a novel normalized matrix attention transformer (Tensorformer) to perform high-quality reconstruction. The proposed matrix attention allows for simultaneous point-wise and channel-wise message passing, while the previous vector attention loses neighbor point information across different channels. It brings more degree of freedom in feature learning and thus facilitates better modeling of local geometries. Our method achieves state-of-the-art on two commonly used datasets, ShapeNetCore and ABC, and attains 4% improvements on IOU on ShapeNet. Code can be accessed https://github.com/THHHomas/Tensorformer6.