Lossless-INR: Lossless Volumetric Implicit Neural Representations

2026-07-20

AI Summary

This work addresses the information loss inherent in existing implicit neural representations for 3D scientific volume reconstruction, which compromises rendering fidelity and downstream analysis. The authors propose a lossless implicit representation by decomposing voxel values into binary bitplanes and formulating reconstruction as a per-bit binary classification task. Their approach integrates adaptive octree partitioning with a triplane feature grid network to enable efficient and accurate modeling. This method achieves, for the first time within an implicit neural representation framework, zero bit error rate and bit-exact reconstruction. It simultaneously supports high-fidelity direct volume rendering, reliable quantitative analysis, and compact storage across diverse volumetric datasets, significantly advancing both reconstruction accuracy and representation efficiency.

Abstract

Implicit neural representation (INR) methods provide continuous coordinate-to-value mappings and integrate naturally with direct volume rendering, making them attractive for representing volumetric data. However, existing INR-based approaches for volumetric data are inherently lossy, and even small reconstruction errors can propagate through rendering and downstream analysis. In this work, we explore Lossless-INR, a lossless INR framework for 3D scientific volumetric data based on bit-plane decomposition. By decomposing each voxel value into binary bit-planes, we reformulate reconstruction as per-bit binary classification, so that exact recovery reduces to predicting every bit correctly. To make this optimization tractable while keeping the representation compact, we combine an octree block-partitioning strategy that adaptively subdivides complex regions with a ternary feature-grid network whose grid entries are parameterized by a ternary set of values. Experiments on diverse volumetric datasets show that this design can achieve zero bit-error rate and bit-exact reconstruction, enabling faithful rendering and downstream analysis with a compact representation. The code is available at https://github.com/TouKaienn/Lossless-INR.

Problem

• lossless compression
• implicit neural representation
• volumetric data
• bit-exact reconstruction
• scientific visualization

Innovation

• Lossless reconstruction
• Implicit Neural Representation
• Bit-plane decomposition
• Octree partitioning
• Ternary feature grid