Optimizing Monocular Depth Estimation with TensorRT: Model Conversion, Inference Acceleration, and 3D Reconstruction
A full-stack AR photogrammetry pipeline that reconstructs real-world objects into 3D models using COLMAP and serves them as interactive, AR-ready GLB files via a FastAPI web server. Captured images are filtered, reconstructed into a sparse point cloud, meshed, colorized, and exported for viewing in-browser with 360° rotation. Scan a QR code .
Create 3D side-by-side and red/cyan anaglyph videos and images from any 2D source.
Camera Calibration; Image Plane to Real world back-projection when the plane which the object lies is known, with respect to the world coordinate system.✔🔴 Note: The most accurate and latest version of the algorithms and documentations are located in branch "M-CAIT-2013"
A pipeline to convert 2d content into 3d stereo side by side video using VideoDepthAnything and python
To associate your repository with the 2dto3d topic, visit your repo's landing page and select "manage topics."