cd /news/computer-vision/data-driven-variable-length-compress… · home › topics › computer-vision › article
[ARTICLE · art-171544] src=arxiv.org ↗ pub= topic=computer-vision verified=true sentiment=· neutral

Data-Driven Variable-Length Compression of 3D Representations

A paper submitted to arXiv on 16 Nov 2026 proposes a learned variable-length compression method for 3D shapes that exploits structure across a training dataset while adapting compressed representation size to each shape's complexity using only fast feed-forward inference. The method, which operates on general latent representations, improves rate-distortion for both meshes and signed distance fields, according to the authors, and code is available at github.com/voilalab/varlen-3D. The work targets a gap in existing 3D compression methods, which either cannot leverage dataset-level statistics or apply a fixed compression ratio to all shapes.

key takeaway

A learned variable-length 3D compression method that uses dataset-level statistics and feed-forward inference improves rate-distortion over fixed-ratio approaches for both meshes and signed distance fields.

by read1 min views3 publishedNov 18, 2026
Data-Driven Variable-Length Compression of 3D Representations
Image: source
  [Submitted on 16 Nov 2026]


[View PDF](https://arxiv.org/pdf/2611.21802)

[HTML (experimental)](https://arxiv.org/html/2611.21802v1)

Abstract:Compressing 3D shapes is important for the efficiency of many applications, yet existing methods either cannot leverage dataset-level statistics or use a fixed compression ratio for all shapes. We propose a learned variable-length compression method that exploits structure across a training dataset while adapting compressed representation size to individual shape complexity using only fast feed-forward inference. Our method operates on general latent representations and improves rate-distortion for both meshes and signed distance fields. Code is available at this https URL.

Additional Features

References & Citations

...

Bibliographic Explorer

(What is the Explorer?) Connected Papers

(What is Connected Papers?) Litmaps

(What is Litmaps?) scite Smart Citations

(What are Smart Citations?) alphaXiv

(What is alphaXiv?) CatalyzeX Code Finder for Papers

(What is CatalyzeX?) DagsHub

(What is DagsHub?) Gotit.pub

(What is GotitPub?) Hugging Face

(What is Huggingface?) ScienceCast

(What is ScienceCast?) Influence Flower

(What are Influence Flowers?) CORE Recommender

(What is CORE?) arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

── more in #computer-vision 4 stories · sorted by recency
── more on @arxiv 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/data-driven-variable…] indexed:0 read:1min 2026-11-18 · —