OmniLatent jointly models geometry, deformation, appearance, and part identities in a shared latent space to generate and interactively edit part-aware 3D assets from images or point prompts.
3D part decomposition is a central capability for editable asset creation, yet existing workflows typically segment a completed mesh as a post-processing step, making the result sensitive to geometric artifacts, texture ambiguity, and limited part-level annotations. We present OmniLatent, a unified latent modeling framework for 3D part decomposition. Instead of predicting part labels in isolation, OmniLatent learns a shared latent space that jointly encodes geometry, deformation, appearance, and binary part IDs, allowing data-rich geometric and visual priors to support data-scarce part learning. Given a single input image, the base model generates geometry, deformation, texture, and…
latent modeling frameworkshared latent spacelatent tokenspretrained modelpoint-prompt adapterbinary part IDsfull segmentationpart subdivisionamodal part completiondynamic per-part token allocation
- Upvotes
- 4
- Authors
- 9
- Published
- Nov 6, 2026
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.