Neural Dynamic GI: Random-Access Neural Compression for Temporal Lightmaps in Dynamic Lighting Environments

2026-04-14

AI Summary

Achieving high-quality global illumination for static scenes under dynamic lighting typically requires precomputing extensive lightmaps, incurring substantial storage and memory overhead. This work proposes a neural compression method tailored for temporal lightmap atlases, uniquely integrating lightweight neural networks with block compression (BC)-mimicking training. By implicitly encoding time-varying illumination into multidimensional feature maps and incorporating a virtual texture (VT) system, the approach enables efficient random access and real-time decompression. The method significantly reduces storage and memory consumption while preserving high-fidelity dynamic global illumination, with decompression overhead remaining well within practical limits.

Technology Category

• Data Mining & Knowledge Management: Data Compression
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures

Abstract

High-quality global illumination (GI) in real-time rendering is commonly achieved using precomputed lighting techniques, with lightmap as the standard choice. To support GI for static objects in dynamic lighting environments, multiple lightmaps at different lighting conditions need to be precomputed, which incurs substantial storage and memory overhead. To overcome this limitation, we propose Neural Dynamic GI (NDGI), a novel compression technique specifically designed for temporal lightmap sets. Our method utilizes multi-dimensional feature maps and lightweight neural networks to integrate the temporal information instead of storing multiple sets explicitly, which significantly reduces the storage size of lightmaps. Additionally, we introduce a block compression (BC) simulation strategy during the training process, which enables BC compression on the final generated feature maps and further improves the compression ratio. To enable efficient real-time decompression, we also integrate a virtual texturing (VT) system with our neural representation. Compared with prior methods, our approach achieves high-quality dynamic GI while maintaining remarkably low storage and memory requirements, with only modest real-time decompression overhead. To facilitate further research in this direction, we will release our temporal lightmap dataset precomputed in multiple scenes featuring diverse temporal variations.

Problem

• global illumination
• dynamic lighting
• lightmap compression
• real-time rendering
• temporal lightmaps

Innovation

• Neural Compression
• Temporal Lightmaps
• Dynamic Global Illumination
• Block Compression Simulation
• Virtual Texturing