G-SHARP: Gaussian Surgical Hardware Accelerated Real-time Pipeline

TL;DR

G-SHARP is a real-time, high-fidelity surgical scene reconstruction framework built on Gaussian splatting, enabling deployable, accurate 3D modeling for minimally invasive surgeries with practical hardware integration.

Contribution

It introduces the first surgical pipeline based on the GSplat differentiable rasterizer, enhancing deformable tissue modeling and occlusion handling for intra-operative use.

Findings

Achieves state-of-the-art reconstruction quality.

Demonstrates real-time performance on edge hardware.

Provides a deployable SDK for surgical visualization.

Abstract

We propose G-SHARP, a commercially compatible, real-time surgical scene reconstruction framework designed for minimally invasive procedures that require fast and accurate 3D modeling of deformable tissue. While recent Gaussian splatting approaches have advanced real-time endoscopic reconstruction, existing implementations often depend on non-commercial derivatives, limiting deployability. G-SHARP overcomes these constraints by being the first surgical pipeline built natively on the GSplat (Apache-2.0) differentiable Gaussian rasterizer, enabling principled deformation modeling, robust occlusion handling, and high-fidelity reconstructions on the EndoNeRF pulling benchmark. Our results demonstrate state-of-the-art reconstruction quality with strong speed-accuracy trade-offs suitable for intra-operative use. Finally, we provide a Holoscan SDK application that deploys G-SHARP on NVIDIA IGX Orin and Thor edge hardware, enabling real-time surgical visualization in practical operating-room settings.