A computer vision project that processes video frames to detect specific features and overlays a 3D model onto the frames based on detected positions. o Extracted frames from video at a specified frame rate using OpenCV. o Loaded and prepared a 3D model (in STL format) with Open3D, including scaling and normalization. o Detected mouth positions in video frames using Mediapipe for accurate feature localization. o Calculated the midpoint between detected mouth positions to determine the overlay position. o Created a 3D scene with Pyrender, positioning the 3D model based on mouth midpoint coordinates. o Rendered the 3D model onto video frames and blended it with the original frames for visual consistency. o Compiled processed frames into a new video with overlaid 3D model using OpenCV.