In Depth We Trust: Reliable Monocular Depth Supervision for Gaussian Splatting
This work addresses the challenge of enhancing 3D reconstruction for computer vision applications using cost-effective depth estimation, though it is incremental as it builds on existing Gaussian Splatting methods.
The paper tackled the problem of unreliable monocular depth priors degrading 3D Gaussian Splatting rendering by introducing a training framework that integrates and selectively regularizes noisy depth inputs, resulting in consistent improvements in geometric accuracy and rendering quality across diverse datasets.
Using accurate depth priors in 3D Gaussian Splatting helps mitigate artifacts caused by sparse training data and textureless surfaces. However, acquiring accurate depth maps requires specialized acquisition systems. Foundation monocular depth estimation models offer a cost-effective alternative, but they suffer from scale ambiguity, multi-view inconsistency, and local geometric inaccuracies, which can degrade rendering performance when applied naively. This paper addresses the challenge of reliably leveraging monocular depth priors for Gaussian Splatting (GS) rendering enhancement. To this end, we introduce a training framework integrating scale-ambiguous and noisy depth priors into geometric supervision. We highlight the importance of learning from weakly aligned depth variations. We introduce a method to isolate ill-posed geometry for selective monocular depth regularization, restricting the propagation of depth inaccuracies into well-reconstructed 3D structures. Extensive experiments across diverse datasets show consistent improvements in geometric accuracy, leading to more faithful depth estimation and higher rendering quality across different GS variants and monocular depth backbones tested.