SAM 3D
Turn any image into 3D objects, people, and scenes — the new generation of Segment Anything AI models for 3D reconstruction.
- Pricing:
- Free
Overview
SAM 3D is a new generation of Segment Anything AI models that reconstruct 3D objects, people, and scenes from a single image. With 300K+ developers using SAM 3D and 1M+ 3D objects reconstructed, it's the breakthrough for single-image 3D reconstruction with zero-shot detection.
SAM3D is a family of AI models released by Meta as part of the Segment Anything collection. While earlier SAM models focused on 2D segmentation, SAM3D takes a step further: it predicts 3D shape, texture, and spatial layout of objects and people directly from a single input image.
Two powerful tools: SAM 3D Objects is a foundation model for reconstructing the full 3D shape, material, and layout of objects and entire scenes from one image, even with clutter or occlusion — providing scene-level understanding and reasoning about multiple objects simultaneously. SAM 3D Body delivers high-accuracy 3D human body pose and shape estimation from a single photo, capturing detailed body geometry and pose — perfect for virtual characters, avatars, and motion analysis.
Classical 3D reconstruction pipelines often require multi-view capture, specialized scanners, or heavy optimization. SAM3D changes the workflow with single-image 3D, scene-level understanding, and real-world robustness. Create 3D product models from photos, segment anything in a scene and 3D-fy it, and use SAM 3D AI as an API with export-ready 3D formats and batch automation support.
Core capabilities
- Single-image 3D
- Reconstruct 3D objects, people, and scenes from one image
- Zero-shot detection
- Segment anything in a scene and 3D-fy it
- SAM 3D Objects
- Scene-level 3D reconstruction with clutter and occlusion robustness
- SAM 3D Body
- High-accuracy 3D human body pose and shape estimation
- 3D product models
- Create export-ready 3D models from photos
- API access
- Use SAM 3D AI as an API with batch automation
- Export-ready formats
- Output 3D in formats ready for games and products
- Meta foundation model
- New generation of Segment Anything AI models
Use cases
- 3D product modeling
- Character creation
- Scene reconstruction
- Avatar generation
- Motion analysis
Who it's for
Game developers, 3D artists, Product designers, Developers, Virtual reality creators
Traffic analytics
sam-3d.comRelated tools
SloydFreeUnlimited AI 3D creation from text, images, or parametric templates, with game-ready output, auto-rigging, and browser-based control.
Hi3DFree planHi3D is an AI-powered, all-in-one 3D creation platform that converts text or reference images into high-fidelity, production-ready 3D models with industry-first 2048³ resolution, 8K PBR textures, and seamless print preparation.
CartwheelPaidCartwheel delivers professional performance-capture animation with AI — multi-character, expressive faces, hands and camera moves, plus text-to-motion, character generation and a motion library for production-ready 3D animation.
Trellis 2PaidAI-powered image-to-3D conversion that turns any 2D image into high-quality 3D models in minutes.