OnnxOCR
基于PaddleOCR重构,并且脱离PaddlePaddle深度学习训练框架的轻量级OCR,推理速度超快 —— A lightweight OCR system based on PaddleOCR, decoupled from the PaddlePaddle deep learning training framework, with ultra-fast inference speed.
- 1.9k
- Python
- Apache-2.0
- 5 mo ago
基于PaddleOCR重构,并且脱离PaddlePaddle深度学习训练框架的轻量级OCR,推理速度超快 —— A lightweight OCR system based on PaddleOCR, decoupled from the PaddlePaddle deep learning training framework, with ultra-fast inference speed.
Pre-trained Deep Learning models and demos (high quality and extremely fast)
Android Voice Activity Detection (VAD) library. Supports WebRTC VAD GMM, Silero VAD DNN, Yamnet VAD DNN models.
Run PyTorch models in the browser using ONNX.js
A set of simple tools for splitting, merging, OP deletion, size compression, rewriting attributes and constants, OP generation, change opset, change to the specified input order, addition of OP, RGB to BGR conversion, change batch size, batch rename of OP, and JSON convertion for ONNX models.
Suite for Windows with Real-ESRGAN, RealESRNet, RealESRAnime, BSRGAN , IRCNN, GFPGAN & RIFE. Upscaling, face restoration, frame interpolation, denoising, batch processing & GPU acceleration in one tool.
Serving AI/ML models in the open standard formats PMML and ONNX with both HTTP (REST API) and gRPC endpoints
Full version of wav2lip-onnx including face alignment and face enhancement and more...
Pre-trained image models using ONNX for fast, out-of-the-box inference.
✨ A real-time voice changer application using RVC, WebSockets and ONNX/TensorFlow/PyTorch
Age and gender estimation from face images in Python. One pip install: no compiler, pretrained dlib models bundled and running on ONNX Runtime. Works with any face detector: OpenCV, MediaPipe, face_recognition.
YOLOv7 to detect bone fractures on X-ray images
🦉Gracefully face reCAPTCHA challenge with ModelHub embedded solution.
Crosswalks Detection using YOLO, project for Computer Vision and Machine Perception course at University of Basilicata, Computer Science and Engineering.
A set of tools and examples for converting and utilizing powerful vision models, DINOv3 and EdgeTAM (SAM2), within the ONNX ecosystem.
Babylon.cpp is a C and C++ library for grapheme to phoneme conversion and text to speech synthesis. For phonemization a ONNX runtime port of the DeepPhonemizer model is used. For speech synthesis VITS models are used. Piper models are compatible after a conversion script is run.
A fast, private, serverless AI background removal (ONNX models) that runs 100% in the browser.
A mobile and ML project for plant analysis, including disease detection, species classification, and leaf/age analysis. Provides Python backend for model training and inference, and a React Native mobile app for on-device usage.
A Flutter mobile application that performs interactive, on-device image segmentation using EdgeTAM (SAM2) encoder/decoder exported to ONNX
Face recognition & embedding framework for Python. Detect, align, and compare faces with LVFace ONNX models — multi-face search, group-photo matching, album clustering. No PyTorch.
ONNX model files (distilbart-mnli zero-shot NLI) for SERP to Prompt Writer, a Windows SEO prompt-generation tool
Pretrained XFeat(Accelerated Features for Lightweight Image Matching) MATLAB Model
A collection of Colab Notebooks designed to export the Google T5Gemma 2 model to ONNX format.
A .NET 10 CLI utility for LLM prompt compression using Microsoft's LLMLingua-2 to hit target token budgets.
Real-time webcam foreground segmentation and background replacement using MODNet, ONNX Runtime, and OpenCV.