IRIS: Intersection-aware Ray-based Implicit Editable Scenes

2026-03-16

AI Summary

This work proposes an efficient implicit scene representation framework that overcomes the limitations of existing methods combining neural radiance fields with 3D Gaussian splatting, which rely on stochastic volumetric sampling and suffer from low rendering efficiency, hindering real-time interactive editing. By leveraging analytical ray-primitive intersection sampling and continuous feature interpolation along rays, the proposed approach eliminates unnecessary empty-space computations and costly 3D neighborhood searches. Integrating 3D Gaussian-guided neural field rendering, the method preserves geometric consistency while significantly accelerating rendering performance, thereby enabling high-fidelity real-time visualization and flexible interactive scene editing.

Technology Category

• Computer Vision: 3D Computer Vision
• Search and Optimization: Sampling/Simulation-based Search
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Neural Radiance Fields achieve high-fidelity scene representation but suffer from costly training and rendering, while 3D Gaussian splatting offers real-time performance with strong empirical results. Recently, solutions that harness the best of both worlds by using Gaussians as proxies to guide neural field evaluations, still suffer from significant computational inefficiencies. They typically rely on stochastic volumetric sampling to aggregate features, which severely limits rendering performance. To address this issue, a novel framework named IRIS (Intersection-aware Ray-based Implicit Editable Scenes) is introduced as a method designed for efficient and interactive scene editing. To overcome the limitations of standard ray marching, an analytical sampling strategy is employed that precisely identifies interaction points between rays and scene primitives, effectively eliminating empty space processing. Furthermore, to address the computational bottleneck of spatial neighbor lookups, a continuous feature aggregation mechanism is introduced that operates directly along the ray. By interpolating latent attributes from sorted intersections, costly 3D searches are bypassed, ensuring geometric consistency, enabling high-fidelity, real-time rendering, and flexible shape editing. Code can be found at https://github.com/gwilczynski95/iris.

Problem

• Neural Radiance Fields
• 3D Gaussian splatting
• ray marching
• feature aggregation
• real-time rendering

Innovation

• analytical sampling
• continuous feature aggregation
• ray-primitive intersection
• real-time rendering
• implicit scene editing