A Query-Efficient Stochastic Volume Rendering Framework for Time-Varying Implicit Neural Volumes

2026-07-30

AI Summary

This work addresses the challenge of interactive rendering of time-varying implicit neural volumes, which is typically hindered by the high computational cost of frequent neural network queries that preclude real-time performance. The authors propose a query-efficient stochastic volume rendering framework based on delta tracking, featuring a four-stage heterogeneous parallel pipeline that drastically reduces the number of neural evaluations. By integrating ray budgeting with query pruning and directly supporting continuous-time exploration, the system enables interactive visualization across temporal domains. It leverages RT cores for ray traversal and tensor cores for accelerated batched neural inference, synergistically combining heterogeneous computing with stochastic rendering. Implemented on an RTX 4090 GPU, the method achieves high-fidelity rendering at 1024×1024 resolution with 30–40 FPS, while switching between timesteps incurs only 1–2 milliseconds of latency.

Abstract

Time-varying implicit neural representations (INRs) provide a compact representation of scientific volumes and, for modalities such as dynamic X-ray computed tomography (CT), are often the only practical way to represent the data. However, interactive volume rendering of INRs is challenging, as cheap memory lookups are replaced by expensive neural inferences, hindering the performance. Therefore, conventional volume rendering methods such as ray marching with dense sampling are often impractical. While resampling, caching, and retraining can mitigate this cost, they compromise convenience and accuracy and become impractical for time-varying data. We tackle these challenges using a query-efficient stochastic volume rendering framework based on delta tracking. Our system employs a four-stage pipeline that exploits heterogeneous parallelism, using ray tracing cores for traversal and tensor cores for batched neural evaluation. Furthermore, we present strategies to reduce INR queries via ray budgeting and query pruning, thereby increasing per-frame performance. Using our renderer, many time-varying INRs can be rendered directly from their original representation. The system achieves ~30-40 FPS at 1024x1024 resolution on an RTX 4090 GPU and converges to high-fidelity images. Moreover, the system enables interactive temporal exploration of the continuous domain, with timestep updates taking approximately 1-2 ms.

Problem

• time-varying implicit neural volumes
• volume rendering
• query efficiency
• interactive visualization
• neural inference

Innovation

• stochastic volume rendering
• implicit neural representations
• delta tracking
• heterogeneous parallelism
• query pruning