Drexubery/ViewCrafter

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1587 stars · 56 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

ViewCrafter is a research codebase that uses video diffusion models to synthesize high-fidelity novel views of scenes from a single or sparse reference images, with precise camera pose control. It supports iterative view synthesis and can feed generated views into 3D Gaussian Splatting optimization for immersive rendering and text-to-3D applications.

Use cases

• generate novel views of a scene from a single image
• synthesize camera trajectory videos with pose control
• create training views for 3D gaussian splatting from sparse images
• scene-level text-to-3D content generation
• zero-shot novel view synthesis research
• reconstruct 3D scenes without dense multi-view captures

When to choose

• you need novel view synthesis from one or a few images without training a per-scene model
• you want precise camera pose control over generated views
• you need generated views to bootstrap 3D-GS optimization
• you are doing research on diffusion-based 3D generation

When to avoid

• you need real-time interactive view synthesis out of the box
• you lack a GPU or cannot run large diffusion models
• you need metrically accurate 3D reconstruction from dense captures
• you want a production-ready application rather than research code

Facets

library · maturity active

machine-learning deep-learning image-processing video-processing computer-vision graphics stable-diffusion computer-vision deep-learning graphics artificial-intelligence image-processing python novel-view-synthesis video-diffusion 3d-reconstruction gaussian-splatting camera-pose-control research-code tpami video gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Drexubery/ViewCrafter")