SCION: Scene Composition with Instanced Neural Primitives

2026-10-01

AI Summary

This study addresses the parameter redundancy and limited editability inherent in existing neural scene representations, which model repetitive elements independently. To overcome these limitations, this work proposes a hierarchical compositional scene representation that replaces independent 3D Gaussians with a reusable primitive library and lightweight instances. Compact and controllable scene reconstruction is achieved through the joint optimization of discrete and continuous parameters, a two-level densification strategy, and adversarial losses. The proposed method significantly reduces storage requirements to merely 1.2MB, outperforming existing approaches while enabling training-free, instance-level editing and animation generation.

Abstract

Real-world scenes are compositional: bricks, blades of grass, pebbles, and tree leaves recur across human-built and natural environments. Existing neural scene representations model these elements independently. Most 3D Gaussian Splatting and follow-up abstraction and compression methods treat each element as unique, fitting millions of independent Gaussians per scene. Prior methods like Splat and Replace fit template objects, but they require mostly manual selection of repeated elements. As a result, these representations store redundant parameters and provide weak manipulation handles for downstream tasks. We introduce SCION, a hier- archical compositional scene representation that replaces independent Gaussians with a compact vocabulary of reusable primitives and lightweight world-space instances that place transformed copies throughout the scene. We fit this represen- tation to multi-view captures via a joint optimization over discrete and continuous scene parameters, combining two-level densification over splats and instances with an adversarial loss that preserves detail across shared primitives. The recovered structure yields a compact, controllable representation while maintaining high quality even at 1.2 MB. SCION achieves rate-distortion favorable to existing Gaussian compression methods, and it enables instance-level scene editing and animation without retraining. Our results show that neural scene representations need not memorize scenes as independent primitives; they can discover reusable parts. Project webpage: https://light.princeton.edu/SCION

Problem

• scene composition
• neural scene representation
• 3D Gaussian Splatting
• parameter redundancy
• instance-level editing

Innovation

• Compositional Scene Representation
• 3D Gaussian Splatting
• Neural Primitives
• Instance-level Editing
• Joint Optimization