I’m building a low-cost, handheld intraoral 3D scanner for everyday clinical use. The hardware is fixed: a Raspberry Pi Camera paired with a simple structured-light projector inside a compact wand. What I still need is the software blueprint and a proof-of-concept workflow that turns a burst of 2-D captures into an accurate printable model. Key objectives • High accuracy on tooth contours—the algorithm must prioritise precise edge and surface definition over colour fidelity. • True portability—everything should run on-board a Pi 4 (or lighter) without calling external cloud services. Minimum deliverables 1. Python/OpenCV script that: – synchronises projector patterns with camera frames, – performs camera calibration, pattern decoding, and robust contour detection, – estimates depth and produces a clean point cloud. 2. A lightweight surface-reconstruction step (Poisson, Marching Cubes, or similar) that outputs an .STL mesh ready for visualisation. 3. Clear, commented code and a short read-me covering dependencies, install steps, and how to trigger a single scan cycle from the command line. Nice-to-have • Live preview of captured frames on the Pi touchscreen. • Simple button-based UI for dentists who aren’t technical. I’m looking for a concise, working prototype rather than a polished commercial product. If you’ve previously handled structured-light, point-cloud generation, or Raspberry Pi imaging, your experience will fit perfectly.