Hugging Face's logo

Spaces:
hugging-apps
/
vega-3d-spatial-reasoning
Running on Zero

App Files Community
vega-3d-spatial-reasoning / README.md
multimodalart's picture
multimodalart HF Staff
Upload folder using huggingface_hub
b5a1a6c verified
preview code
|
Raw
History Blame Contribute Delete
1 kB

A newer version of the Gradio SDK is available: 6.24.0

Upgrade
metadata 
title: VEGA-3D Spatial Reasoning
emoji: 馃Л
colorFrom: red
colorTo: gray
sdk: gradio
sdk_version: 5.49.1
app_file: app.py
short_description: 3D spatial reasoning VLM with video-diffusion 3D priors
python_version: '3.10'
startup_duration_timeout: 1h
pinned: false

VEGA-3D 路 Spatial Reasoning

Interactive demo of VEGA-3D, a spatial-reasoning multimodal LLM from the paper Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understanding (ECCV 2026).

VEGA-3D augments a Qwen2.5-VL backbone with implicit 3D priors extracted from a frozen Wan2.1-T2V-1.3B video diffusion model. Intermediate spatiotemporal features from the diffusion model are fused into the VLM via token-level gated fusion, giving it stronger geometric and spatial understanding of indoor scenes.

Upload a scene image (or a short scene video) and ask a spatial-reasoning question.