← Research Timeline Aditya Jain / Apple Maps · 3D Reconstruction
Jan 2026
Topic 31 Jan 2026 UODF · Neural Implicit · Paper Study

UODF —
Three Axes, Three Distance Fields.

Paper study session on Unsigned Orthogonal Distance Fields (CVPR 2024). The core idea: instead of storing the minimum distance to the surface in any direction (SDF/UDF), store three separate axis-aligned unsigned distance fields — UODF_LR along X, UODF_FB along Y, UODF_UD along Z. The pay-off is that surface extraction needs no marching-cubes interpolation step — each ray's intersection is a direct table lookup along its axis. 20–100× better on open / non-watertight surfaces; the triplane-variant runs 30–80 M points/sec on a consumer GPU. Informs the six-plane-mesh and hexplane work in the thesis line.

00 — Motivation

Marching cubes' interpolation error kills open-surface reconstruction.

Standard neural-implicit surface reconstruction (SDF + marching cubes) has two error sources: fitting error (the network's distance estimate is imperfect) and interpolation error (marching cubes linearly interpolates between corner-grid distance values to find the zero-crossing, which is inaccurate for non-linear surfaces). For watertight closed surfaces both errors are small. For open surfaces (garments, leaves, single-sided geometry) or for complex internal structure (assemblies, hollow objects, hexplane-style multi-layer surfaces) the interpolation error dominates and reconstruction quality collapses.

The UODF paper (Lu et al., CVPR 2024) sidesteps the marching-cubes interpolation step entirely. By storing the unsigned distance along a fixed axis direction (not "in any direction"), each grid cell's distance value is the surface intersection — no interpolation needed. The catch: you have to store three of them (one per axis) because a single axis-aligned UODF only covers rays in that direction.

What it informs
The UODF approach is the natural-feeling generalisation of the six-plane-mesh thesis topic (Topic 32, single axis-aligned direction per plane × 6 planes = covers the geometry from all orthographic angles) and the hexplane-AE topic (Topic 29, 6-view feature representation of a 3-D mesh). Both use the same axis-aligned principle UODF formalises.
01 — The Representation

Three orthogonal unsigned distance fields, one per axis.

FieldStoresQuery semantics
UODF_LR (X-axis)Distance to nearest surface along ±X from each grid voxel"How far to the surface if I shoot a ray along X?"
UODF_FB (Y-axis)Distance along ±YSame, Y
UODF_UD (Z-axis)Distance along ±ZSame, Z

Surface extraction: for each grid voxel, the three UODF values name three candidate surface points along three axes; the union of these points across all voxels is the reconstructed surface. The crucial difference from marching cubes: no interpolation is needed because each candidate point is the actual axis-aligned distance to the surface, not a corner-value interpolation.

SDF + marching cubes : N³ voxels → interpolate 8 corners → triangle UODF (3 axis fields) : N³ voxels → 3 direct surface points each → union Triplane UODF (fast variant): 3 × N² 2-D planes → bilinear sample → direct surface Speed (paper-reported): SDF + marching cubes : ~0.3–1 M points/sec on RTX 3060 UODF : ~5–10 M points/sec Triplane UODF : 30–80 M points/sec
Core Insight

Axis-aligned distance ≠ minimum distance.
The constraint is the feature.

The SDF / UDF approach throws away axis information — the stored distance is to the nearest surface in any direction. Recovering a surface from that means interpolating gradients (marching cubes). UODF keeps the axis information by storing three separate fields per voxel. The constraint (must commit to an axis up-front) is what makes surface extraction interpolation-free.

Interactive Demo · Live

Click ray-cast directions on the left pane to see each UODF field separately. The middle pane shows the union of all three axis-fields — the reconstructed surface. The right pane shows the same shape reconstructed via SDF + marching cubes for comparison, with the interpolation error visible at low resolutions.

01 — Active axis field UODF_LR (X)
02 — Three-axis union SPHERE
03 — SDF + marching cubes interpolation error visible

Full Technical Paper

White paper · UODF paper analysis · axis-aligned principle · thesis-line cross-reference to Six-Plane Mesh and Hexplane AE

Read Paper →
Related Thesis Chapters
Six-Plane Mesh Reconstruction
Same axis-aligned-distance idea, applied to depth-map → mesh extraction with six planes (one per ±axis).
Hexplane Autoencoder
Same six-view representation, applied to VAE / AE training instead of distance-field storage.
Triplane Deep Dive
The triplane-UODF variant in the paper is the same triplane scaffold studied separately in Topic 30.
Appendix — Raw Materials
Transcripts & Source References
████████████████████████████████████████████████

██████████████████████████████████████
█████████ · ████ · █████████████████████
████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████
Restricted Access