Inference Focused Library - Onnx Models -> IR -> Hardware Instructions

YOLOv3 in PyTorch > ONNX > CoreML > iOS

Folder with code related to object detection in the CCTV cameras placed in the agricultural field and also down streaming for agricultural use-case

Tools for ONNX models

Deep learning-based semantic segmentation model using the PyTorch and ONNX framework

ML-driven snake species classification

Production-ready distributed YOLO inference pipeline powered by NVIDIA Triton Inference Server. Supports Kubernetes orchestration and Docker deployment.

Inspired by checkout line assignments, this project simulates advanced load balancing in cloud systems, using reinforcement learning to assign jobs with uncertain workloads across servers.

JVM library for voice activity detection written in Kotlin based on C library fvad and Silero

Object detection model builder with TensorFlow

A Streamlit (Python Web Framework) Application that detects most common human activities from Pre-Recorded Videos or Live Camera Feed.

A multi-label text classifier model which can suggest cover page colors based on the description(with/without title) and genres of a particular book. The predicted set of colors (10 colors) can also be used to generate multi-color Barcode for the book, where we can use the prediction % as the width of the bars.

People detection using YOLOv8 with support for video, RTSP, webcam, and image datasets. Includes model selection, optimization (SSIM, motion detection), multiprocessing, and quantization-ready ONNX models.

Python side of training model for Gradient boosting regressor converted to onnx format + MLP classifier training.

This is a modified version of YOLOv5 popular object detection model that mainly works for Intel GPUs that accelerates the video inference on Iris(X) GPUs and other intel family GPUs mainly for the Video Processing activities feel free to give the issue

PFLD - Ultralight - 400FPS(CPU) 98关键点, 可以用来训练106个特征点

MaskRCNN-Pytorch provides a comprehensive C++ development workflow designed to help you train models, convert ONNX format, and C++ deploy.

ONNX format for ferdous-alam/GenCAD model.

Create your own Sign language and then translate it live.

A complete pipeline involving image processing techniques and machine learning models for distinguishing civilian and military vehicles across multiple types, supporting both binary and multilabel classification, and optimized for real-time edge deployment using ONNX and TensorRT.