Learning Fine-Grained Geometry for Sparse-View Splatting via Cascade Depth Loss

2025-05-28

AI Summary

Novel view synthesis from sparse views suffers from structural distortions and blurred details due to insufficient geometric cues, severely limiting NeRF- and 3D Gaussian Splatting (3DGS)-based methods. To address this, we propose Hierarchical Depth-Guided Splatting (HDGS), a novel framework that leverages monocular depth estimation as a geometric prior to enforce multi-scale depth consistency. HDGS introduces a Cascaded Pearson Correlation Loss (CPCL) that regularizes geometry via cross-scale feature correlation, enabling correlation-driven geometric regularization. Built upon 3D Gaussian Splatting, HDGS integrates depth guidance, multi-scale rendering supervision, and explicit geometric constraints—significantly improving reconstruction accuracy and rendering fidelity. Extensive experiments on the LLFF and DTU benchmarks under sparse-view settings demonstrate state-of-the-art performance, with substantial gains in geometric fidelity and fine-detail preservation.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Novel view synthesis is a fundamental task in 3D computer vision that aims to reconstruct realistic images from a set of posed input views. However, reconstruction quality degrades significantly under sparse-view conditions due to limited geometric cues. Existing methods, such as Neural Radiance Fields (NeRF) and the more recent 3D Gaussian Splatting (3DGS), often suffer from blurred details and structural artifacts when trained with insufficient views. Recent works have identified the quality of rendered depth as a key factor in mitigating these artifacts, as it directly affects geometric accuracy and view consistency. In this paper, we address these challenges by introducing Hierarchical Depth-Guided Splatting (HDGS), a depth supervision framework that progressively refines geometry from coarse to fine levels. Central to HDGS is a novel Cascade Pearson Correlation Loss (CPCL), which aligns rendered and estimated monocular depths across multiple spatial scales. By enforcing multi-scale depth consistency, our method substantially improves structural fidelity in sparse-view scenarios. Extensive experiments on the LLFF and DTU benchmarks demonstrate that HDGS achieves state-of-the-art performance under sparse-view settings while maintaining efficient and high-quality rendering

Problem

• Improving sparse-view 3D reconstruction quality
• Addressing blurred details in Neural Radiance Fields
• Enhancing geometric accuracy via multi-scale depth consistency

Innovation

• Hierarchical Depth-Guided Splatting (HDGS) framework
• Cascade Pearson Correlation Loss (CPCL)
• Multi-scale depth consistency enforcement