DistScene: Object-to-Scene Distillation for 3D Scene Generation
DistScene generates compositional 3D scenes from one image by jointly modeling environments and objects.
DistScene generates compositional 3D scenes from a single image by modeling the environment as an explicit component rather than only a collection of objects. Scene-Frame Generation places environment and object geometry in a shared coordinate frame, Object-Centric Refinement improves each object locally, and Object-to-Scene Distillation transfers pretrained object priors using synthetic rendered scenes. Indoor and outdoor evaluations report better scene-level spatial coherence than the compared baselines.
- Models environment geometry explicitly alongside individual objects.
- Generates scene and object components in a shared coordinate frame.
- Distills pretrained object-generation priors into scene generation.
- Improves spatial coherence on indoor and outdoor benchmarks.
From the original article
1 min read at huggingface.co
We present DistScene, a framework for single-image compositional 3D scene generation by jointly modeling the environment and individual objects. Unlike existing methods that represent scenes primarily as collections of objects, we model the environment as an explicit scene component to provide geometric context for object placement.
By Kunming Luo, Hongyu Yan, Ken Deng · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.