IRIS: Intersection-aware Ray-based Implicit Editable Scenes

TL;DR

IRIS is a novel framework that enables efficient, real-time scene rendering and editing by analytically identifying ray-scene interactions and aggregating features directly along rays, combining neural radiance fields' quality with Gaussian splatting's speed.

Contribution

It introduces an intersection-aware, ray-based implicit scene representation that significantly improves rendering efficiency and editing flexibility over existing methods.

Findings

Achieves real-time rendering with high fidelity.

Enables flexible scene editing.

Eliminates costly volumetric sampling.

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.