Awesome GitHub RepositoriesTensorFlow.js Models

TensorFlow.js is a JavaScript machine learning library used for training and deploying models in web browsers and server-side environments. It functions as a browser-based model trainer, a WebAssembly inference engine, and a WebGPU accelerated tensor library for low-level linear algebra. The project also includes a model converter to transform Python-based models into optimized formats for JavaScript execution. The library distinguishes itself through a pluggable backend architecture that allows mathematical operations to be executed via CPU, WebGL, or WebGPU. It supports the conversion of Py

NSFW detection on the client-side via TensorFlow.js

Kalidokit is a web-based motion capture tool that transforms real-time webcam video into 3D character animation data. It functions as a blendshape and kinematics calculator, converting facial, hand, and body tracking data from Mediapipe and TensorFlow.js into blendshape weights and euler rotations for driving digital puppets and avatars. The tool solves face landmarks to derive head rotation, eye blinks, mouth shapes, and brow values for rigging, while hand landmarks are converted into finger joint rotations and body keypoints into per-joint euler rotations for full-body animation. It include

Real-Time-Person-Removal este o aplicație de computer vision bazată pe web, concepută pentru a identifica și elimina figurile umane din fluxurile video live. Folosind TensorFlow.js, instrumentul funcționează ca un sistem de scădere a fundalului în timp real care analizează compoziția scenei pentru a izola fundalurile statice de persoanele în mișcare. Proiectul permite computer vision bazat pe browser prin procesarea fluxurilor video de la webcam direct în client. Utilizează machine learning pentru a diferenția între elementele dinamice ale scenei și fundal, permițând eliminarea în timp real a persoanelor din câmpul vizual.

This project is a JavaScript library designed for real-time face detection directly within a web browser. It functions as a machine learning model wrapper that enables developers to identify and track human faces in live video streams without the need for backend server processing. The library utilizes browser-native media access to stream raw camera data into application memory, where it performs pixel-level analysis. By leveraging a tensor-based inference engine and web-assembly acceleration, the tool executes complex neural network calculations locally to achieve high-performance computer