WebXR, augmented reality MIDI data visualization, built with Three.js and Tone.js. See video: https://youtu.be/lIecCGtbqSM

Explore the fascinating world of depth perception with this cutting-edge web application! Using the power of TensorFlow.js and the MiDaS (Monocular Depth Estimation) model converted to TensorFlow Lite format, this project brings real-time depth prediction right to your browser.

AI-powered tool for creating advanced visual effects (VFX) in videos

This code provides depth estimation function, while supporting Web deployment, through the Web page to try to single image depth estimation.

Demonstration of real time depth estimation on the browser without depth sensors in Three.Js and A-Frame using machine learning models FastDepth and others.

Assisting visually impaired individuals in navigation through computer vision.

Project Iris is a mobile-first, web-based application built with React and TensorFlow.js, providing real-time on-device object detection and depth estimation. It includes user monitoring, an admin dashboard, and customizable settings.

wip basic node graph system for photometric stereo

This project focuses on enhancing 2D videos to create a 3D-like immersive experience using a combination of deep learning-based instance segmentation and depth estimation techniques.

Add depth estimation to your Blazor WebAssembly project with Transformers.js and Depth Anything V2 Small for Blazor WebAssembly. Includes fp16 and fp32 models.

PowerToys, but AI

Repository for creating a demo streamlit app.