Splat Capture
Point your phone at a real object, walk around it, and get a photorealistic 3D model — trained on your own GPU, ready to 3D print.
Gaussian splatting is a newer 3D reconstruction technique that produces dramatically better results than traditional mesh photogrammetry, especially for organic shapes and fine details. Splat Capture wraps the entire pipeline — photo capture, transfer, SfM, training, viewing, and export — into a single cohesive tool that runs on consumer hardware.
Why Gaussian Splatting
Traditional photogrammetry reconstructs a mesh — it works well for large objects with clear geometric structure but struggles with hair, fur, translucent surfaces, and fine detail. Gaussian splatting represents the scene as millions of 3D Gaussians (ellipsoids), each with its own color, opacity, and orientation. The result is a volumetric representation that captures lighting and fine detail far better than a polygon mesh.
The tradeoff is compute: 30K training iterations on a consumer GPU takes 15–25 minutes. The output isn't directly 3D-printable either — the export step converts the gaussian field to a mesh (STL/3MF) for printing, with some loss in fine detail at the conversion stage.
Pipeline
Launch Desktop App
The Windows app (Python + CustomTkinter) starts a local FastAPI server and displays a QR code. It's your command center for the whole pipeline.
Connect Android App
Scan the QR code with the Splat Capture Android app. It connects over WiFi to the PC, no pairing or Bluetooth required. The app was built with Expo / React Native and distributed via EAS.
Guided Photo Capture
Choose 60, 120, or 200 photo mode. The app guides you around the subject with visual prompts — telling you where to move and when to shoot to ensure good coverage from all angles. More photos = better reconstruction.
WiFi Transfer
Photos stream directly to the PC as you shoot, or batch-transfer at the end. No USB cables, no cloud upload, no AirDrop. They land in a dated session folder ready for processing.
COLMAP Structure-from-Motion
COLMAP runs feature extraction and matching across all photos, then builds a 3D point cloud and estimates camera positions via Structure-from-Motion. This is what tells the gaussian splat trainer where every photo was taken from.
Gaussian Splat Training
30,000 training iterations of gsplat 0.1.12 on your NVIDIA GPU. At this iteration count on a GTX 1060 (sm_61), a typical indoor object trains in roughly 15–25 minutes. The result is a dense 3D gaussian splat — not a mesh, but a volumetric point cloud that renders photorealistically.
WebGL Viewer
The trained splat loads in a WebGL viewer (based on antimatter15/splat) embedded in the desktop app. Orbit, zoom, and inspect the result before committing to export.
Edit & Export
A Three.js splat editor lets you trim unwanted gaussians (background clutter, floating artifacts) by drawing selection boxes. Export the cleaned model to STL or 3MF for 3D printing.