Revisiting Depth Representations for Feed-Forward 3D Gaussian Splatting

TL;DR

This paper introduces PM-Loss, a regularization technique using a transformer-predicted pointmap to improve depth maps in 3D Gaussian Splatting, enhancing rendering quality by addressing depth discontinuities.

Contribution

It proposes a novel regularization loss, PM-Loss, that leverages a transformer-based pointmap to improve depth map quality in 3D Gaussian Splatting.

Findings

Significant improvement in rendering quality across various scenes.

Enhanced geometric smoothness at object boundaries.

Better performance with multiple 3DGS architectures.

Abstract

Depth maps are widely used in feed-forward 3D Gaussian Splatting (3DGS) pipelines by unprojecting them into 3D point clouds for novel view synthesis. This approach offers advantages such as efficient training, the use of known camera poses, and accurate geometry estimation. However, depth discontinuities at object boundaries often lead to fragmented or sparse point clouds, degrading rendering quality -- a well-known limitation of depth-based representations. To tackle this issue, we introduce PM-Loss, a novel regularization loss based on a pointmap predicted by a pre-trained transformer. Although the pointmap itself may be less accurate than the depth map, it effectively enforces geometric smoothness, especially around object boundaries. With the improved depth map, our method significantly improves the feed-forward 3DGS across various architectures and scenes, delivering consistently better rendering results. Our project page: https://aim-uofa.github.io/PMLoss