Haochen-Wang409/ross3d
[ICCV'25] Ross3D: Reconstructive Visual Instruction Tuning with 3D-Awareness
Ross3D helps AI models understand complex 3D environments by enabling them to reconstruct full scenes from various camera angles. It takes in multiple 2D views or video frames of a 3D space and outputs a comprehensive understanding of that scene, like a 'Bird's-Eye-View.' This is useful for AI researchers and developers building systems that need to interpret and interact with physical 3D spaces.
No commits in the last 6 months.
Use this if you are developing AI models that need to accurately understand and interpret information from 3D environments, especially when working with limited 3D data.
Not ideal if your primary focus is on 2D image analysis or if you require an off-the-shelf application rather than a foundational AI model.
Stars
67
Forks
1
Language
Python
License
Apache-2.0
Category
Last pushed
Jul 22, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/Haochen-Wang409/ross3d"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
col14m/cadrille
[ICLR2026] cadrille: Multi-modal CAD Reconstruction with Online Reinforcement Learning
filaPro/cad-recode
[ICCV2025] CAD-Recode: Reverse Engineering CAD Code from Point Clouds
pengsongyou/openscene
[CVPR'23] OpenScene: 3D Scene Understanding with Open Vocabularies
worldbench/3EED
[NeurIPS 2025 DB Track] 3EED: Ground Everything Everywhere in 3D
cambrian-mllm/cambrian-s
Cambrian-S: Towards Spatial Supersensing in Video