A slice-based approach mostly keeps holes and connections in 3D from images
A compact, slice-based 3D representation can keep important topology like holes and connections, but may struggle with some complex shapes.
- Publication
- arXiv
- Stage
- Preprint
- What we read
- Summary of the paper
- Authors
- Tianjiao Yu, Xinzhuo Li, Yifan Shen, Ying Shen, Kiet A. Nguyen, Adheesh Sunil Juvekar, Ismini Lourentzou
- Universities and research institutions
- Not yet supplied in verified metadata; the Brief does not guess.
What they did and found
A way encodes 3D shapes by stacking flat slices along three directions to keep holes and connections. It uses less memory and is faster, but complex shapes may fail.
Why it matters
Keeping holes and connections helps 3D models look right and work in tools. Some thin or intricate parts may still be imperfect.
The approach treats each axis as a stream of flat slices that together describe the 3D shape, helping the model preserve holes and connections.
In testing, you would sample items from a catalog to see if the shape connections hold in typical cases, not just easy examples.
What remains uncertain
Performance depends on how varied the data are; unusual or very thin shapes may not be captured well in all cases.
Original sources · 1
- SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation ↗arXiv · 2026-10-01
Check the original paper for its authors, methods, version and access terms.
What is your take?
Ask a question, add useful context or share a different perspective. Keep the conversation respectful and grounded.