Get Started
Home
Topics
Search
Library
Research questionHow can Gaussian splatting reduce per-primitive appearance memory without limiting angular detail?In 3D Gaussian Splatting, spherical-harmonic coefficients consume substantial per-primitive storage and memory traffic, while their band-limited basis can miss fine view-dependent effects. More expressive representations can also change optimization behavior and absorb non-static scene content, complicating the quality–efficiency trade-off.
AI
Computer Vision
Evaluation & Benchmarks
Image & Video Processing
Inference Optimization
Machine Learning
Neural and Evolutionary Computing
Research Paper
Small / On-device Models
Technology
Latest papersRecent research connected to this question, newest first.Compact Neural Appearance Models for Efficient Gaussian SplattingThe source compares spherical-harmonic and newer spherical appearance models within a common optimized Gaussian-splatting pipeline. Evidence covers reconstruction quality, per-primitive memory, optimization and rendering performance, recovered geometry, and non-static content absorption; the proposed compact alternative uses per-primitive latent codes decoded by a shared tiny MLP, with CUDA rasterization and a WebGL viewer for laptop and mobile GPUs.research paper · Sep 4, 2026
Related questions
How can existing 3D Gaussian Splatting models be compacted for storage and transmission while preserving novel-view fidelity?How can dynamic 4D Gaussian splatting reduce per-scene storage without sacrificing view-synthesis quality or real-time rendering?How can 3D Gaussian Splatting training stay efficient as high-resolution scenes require more Gaussian primitives?How can feed-forward 3D Gaussian Splatting meet a fixed Gaussian budget without sacrificing novel-view quality?