AG$^2$aussian: Anchor-Graph Structured Gaussian Splatting for Instance-Level 3D Scene Understanding and Editing
By: Zhaonan Wang, Manyi Li, Changhe Tu
Potential Business Impact:
Organizes 3D scenes for precise object editing
3D Gaussian Splatting (3DGS) has witnessed exponential adoption across diverse applications, driving a critical need for semantic-aware 3D Gaussian representations to enable scene understanding and editing tasks. Existing approaches typically attach semantic features to a collection of free Gaussians and distill the features via differentiable rendering, leading to noisy segmentation and a messy selection of Gaussians. In this paper, we introduce AG$^2$aussian, a novel framework that leverages an anchor-graph structure to organize semantic features and regulate Gaussian primitives. Our anchor-graph structure not only promotes compact and instance-aware Gaussian distributions, but also facilitates graph-based propagation, achieving a clean and accurate instance-level Gaussian selection. Extensive validation across four applications, i.e. interactive click-based query, open-vocabulary text-driven query, object removal editing, and physics simulation, demonstrates the advantages of our approach and its benefits to various applications. The experiments and ablation studies further evaluate the effectiveness of the key designs of our approach.
Similar Papers
CUS-GS: A Compact Unified Structured Gaussian Splatting Framework for Multimodal Scene Representation
CV and Pattern Recognition
Makes 3D worlds look real with less computer power.
CAGS: Open-Vocabulary 3D Scene Understanding with Context-Aware Gaussian Splatting
CV and Pattern Recognition
Helps robots understand 3D objects from any words.
Beyond Averages: Open-Vocabulary 3D Scene Understanding with Gaussian Splatting and Bag of Embeddings
CV and Pattern Recognition
Lets computers understand and find objects in 3D.