C³-GS: Learning Context-aware, Cross-dimension, Cross-scale Feature for Generalizable Gaussian Splatting

2025/08/28 by Hu, Yuxi, Zhang, Jun, Chen, Kuangyi +2
#Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2508.20754

Abstract

Generalizable Gaussian Splatting aims to synthesize novel views for unseen scenes without per-scene optimization. In particular, recent advancements utilize feed-forward networks to predict per-pixel Gaussian parameters, enabling high-quality synthesis from sparse input views. However, existing approaches fall short in encoding discriminative, multi-view consistent features for Gaussian predictions, which struggle to construct accurate geometry with sparse views. To address this, we propose C³-GS, a framework that enhances feature learning by incorporating context-aware, cross-dimension, and cross-scale constraints. Our architecture integrates three lightweight modules into a unified rendering pipeline, improving feature fusion and enabling photorealistic synthesis without requiring additional supervision. Extensive experiments on benchmark datasets validate that C³-GS achieves state-of-the-art rendering quality and generalization ability. Code is available at: https://github.com/YuhsiHu/C3-GS.

Citations

• Frequency-Aligned Knowledge Distillation for Lightweight Spatiotemporal Forecasting
• SRKD: Towards Efficient 3D Point Cloud Segmentation via Structure- and Relation-aware Knowledge Distillation
• ICG-MVSNet: Learning Intra-view and Cross-view Relationships for Guidance in Multi-View Stereo
• EVLoc: Event-based Visual Localization in LiDAR Maps via Event-Depth Registration
• DiffSSC: Semantic LiDAR Scan Completion using Denoising Diffusion Probabilistic Models
• MVPGS: Excavating Multi-view Priors for Gaussian Splatting from Sparse Input Views
• MVSGaussian: Fast Generalizable Gaussian Splatting Reconstruction from Multi-View Stereo
• Geometry-aware Reconstruction and Fusion-refined Rendering for Generalizable Neural Radiance Fields
• SLCF-Net: Sequential LiDAR-Camera Fusion for Semantic Scene Completion using a 3D Recurrent U-Net
• Splatter Image: Ultra-Fast Single-View 3D Reconstruction
• pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction
• GPS-Gaussian: Generalizable Pixel-wise 3D Gaussian Splatting for Real-time Human Novel View Synthesis
• 3D Gaussian Splatting for Real-Time Radiance Field Rendering
• 3D Gaussian Splatting for Real-Time Radiance Field Rendering
• Explicit Correspondence Matching for Generalizable Neural Radiance Fields
• MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth
• Rethinking Depth Estimation for Multi-View Stereo: A Unified Representation
• TransMVSNet: Global Context-aware Multi-view Stereo Network with Transformers
• GMFlow: Learning Optical Flow via Global Matching
• MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo
• IBRNet: Learning Multi-View Image-Based Rendering
• pixelNeRF: Neural Radiance Fields from One or Few Images
• Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement
• Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching
• Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelines
• MVSNet: Depth Inference for Unstructured Multi-view Stereo
• The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
• Feature Pyramid Networks for Object Detection
• Image quality assessment: from error visibility to structural similarity

Related

• G3Splat: Geometrically Consistent Generalizable Gaussian Splatting
• GS-Net: Generalizable Plug-and-Play 3D Gaussian Splatting Module
• Feature Splatting for Better Novel View Synthesis with Low Overlap
• IBGS: Image-Based Gaussian Splatting
• PointGS: Point Attention-Aware Sparse View Synthesis with Gaussian Splatting