Coding & Development

YoloDotNet is a modular, lightweight C# library built on .NET 8, ONNX Runtime, and SkiaSharp, designed for real-time computer vision and YOLO-based inference. It offers high-performance inference for modern YOLO model families (YOLOv5u through YOLOv26, YOLO-World, YOLO-E, and RT-DETR) without relying on heavy computer vision frameworks like OpenCV or Python runtimes. Developers gain explicit control over execution, memory, and preprocessing, making it ideal for production-ready desktop apps, backend services, and real-time vision pipelines requiring deterministic behavior. It supports various vision tasks including classification, object detection, OBB detection, segmentation, and pose estimation, with flexible execution providers for CPU, CUDA/TensorRT, OpenVINO, CoreML, and DirectML.

YOLOv3 is an open-source Keras implementation of the YOLOv3 object detection algorithm, designed for identifying objects within images and videos. This tool requires specific dependencies including OpenCV 3.4, Python 3.6, TensorFlow-gpu 1.5.0, and Keras 2.1.3. Users can quickly get started by downloading official YOLOv3 weights and converting them to a Keras H5 file using the provided `yad2k.py` script. The tool demonstrates improved classification capabilities over its predecessor, YOLOv2. While it currently supports object detection, future development plans include training the model for broader applications. It is a valuable resource for developers and data scientists working on computer vision tasks.

nvim-luapad is an interactive real-time scratchpad designed for Neovim's embedded Lua engine, enabling developers to quickly test and debug Lua code directly within their editor. It evaluates code in real-time, displaying output as virtual text where it was called. The tool extends the native Lua command with deep function completion, streamlining the development of Neovim plugins. While powerful for small code chunks, users should exercise caution as it evaluates every typed character, which can lead to unintended side effects with system calls or UI manipulations. It offers configurable options like error indicators, preview windows, and evaluation triggers, making it a flexible tool for Neovim Lua development.

Nidium is an ongoing open-source project focused on developing a general-purpose, hardware-accelerated rendering engine for creating both desktop and mobile applications and games. It enables developers to build graphical software using JavaScript, distinguishing itself from NodeJS, QT, Chromium, or WebKit derivatives. Designed from scratch with a small C++ codebase, Nidium integrates powerful libraries like Google Skia for graphics and Mozilla's SpiderMonkey for JavaScript execution. It supports common APIs such as WebGL, Canvas 2D, and WebSockets, alongside non-standard features like UDP/TCP sockets and threaded audio. Nidium also offers a server-side component, nidium-server, sharing the non-graphics codebase for comprehensive application development.

Animatable_nerf is an open-source research tool that provides the implementation for "Animatable Implicit Neural Representations for Creating Realistic Avatars from Videos," a paper accepted to TPAMI 2024 and ICCV 2021. This tool allows researchers to generate realistic avatars from video footage by leveraging animatable neural fields. It supports various configurations, including vanilla Animatable NeRF, versions with neural blend weight fields replaced by displacement fields, and versions where the canonical NeRF model is replaced with a neural surface field (SDF output). The repository includes evaluation frameworks for reconstruction quality comparison and provides access to datasets like Mobile-Stage and SyntheticHuman++ for further research and development in neural rendering and 3D human body modeling.

PyContrast is a PyTorch-based library designed for researchers and practitioners working with contrastive learning methods. It offers a comprehensive collection of recent contrastive learning papers and provides reference implementations for state-of-the-art techniques such as InstDis, CMC, and MoCo. The library also includes a set of pre-trained ImageNet unsupervised models, which can be found in its model zoo. These unsupervised pre-training models have demonstrated superior performance over supervised models in object detection tasks on datasets like PASCAL VOC and COCO, making PyContrast a valuable resource for advancing self-supervised learning research and applications.

blinker-library is a comprehensive IoT solution designed for embedded hardware, offering cross-hardware and cross-platform compatibility. It supports popular microcontrollers like Arduino R4, ESP32, and ESP8266, making it versatile for various IoT applications. The library provides full-stack support, including APP, device, and server components, and leverages public cloud services for efficient data transmission and storage. Inspired by the simplicity of the 'Blink' sketch, blinker aims to make IoT project development as straightforward as lighting an LED. It's ideal for smart home systems, data monitoring, and other IoT fields, helping users build projects more quickly and effectively.

CefSharp is a powerful open-source library that provides .NET bindings for the Chromium Embedded Framework (CEF). This allows developers to seamlessly embed a full-featured Chromium web browser control directly into their Windows Presentation Foundation (WPF) and Windows Forms applications. The library is primarily written in C# with a portion of the bindings in C++/CLI, making it accessible from C#, VB, or any other CLR language. CefSharp is BSD licensed, ensuring it can be used freely in both proprietary and open-source projects. It offers various NuGet packages for different application types, including WinForms, WPF, and OffScreen, along with detailed examples and documentation to assist with integration and feature utilization.

Sniprun is a Neovim plugin written in Lua and Rust designed for quickly running lines or blocks of code directly within the editor. It supports a wide array of interpreted and compiled languages, offering a fast partial code testing experience. The tool aims to provide REPL-like behavior for many languages, allowing users to execute code dependent on previously run snippets. Sniprun is ideal for learning new languages, experimenting with features, or developing simple code pipelines. It boasts features like automatic import fetching, live execution, and various result display modes, making it a versatile tool for developers seeking rapid iteration and experimentation.

SoftTeacher is an open-source project providing the official implementation of the ICCV2021 paper "End-to-End Semi-Supervised Object Detection with Soft Teacher." This tool enables the training of object detection models using a combination of labeled and unlabeled data, significantly improving model accuracy, especially in scenarios with limited labeled data. It offers configurations for various labeled data percentages (1%, 5%, 10%, and full labeled data) and includes pre-trained model weights. The repository provides detailed instructions for installation, data preparation, training, evaluation, and inference, making it a valuable resource for researchers and developers in computer vision.

SRGAN is a PyTorch implementation of the Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network paper from CVPR 2017. This open-source tool allows users to perform super-resolution on both images and videos, significantly enhancing their quality and detail. It provides options for various upscale factors (2x, 4x, 8x) and includes benchmarks for performance on different datasets. Users can train their own models, test on benchmark datasets, or apply super-resolution to single images and videos using pre-trained models. The project is hosted on GitHub and requires Anaconda, PyTorch, and OpenCV for setup.

SRGAN-tensorflow offers a TensorFlow implementation of the SRGAN algorithm, designed for single image super-resolution. This project is based on the impressive work "Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network." It allows users to upscale images, achieving results comparable to those presented in the original research paper, even with limited resources. The tool supports both testing with pre-trained models and training new models on custom datasets like RAISE. It provides scripts for running inference, testing, and training SRResnet and SRGAN models with different perceptual losses (MSE and VGG54). The code is highly inspired by pix2pix-tensorflow and includes detailed instructions for setting up dependencies and executing various modes.

esp-idf-hal is an open-source project that offers safe Rust wrappers for the drivers found in the ESP IDF SDK, specifically designed for ESP32 microcontrollers. It implements both V0.2 and V1.0 traits of embedded-hal, supporting both blocking and asynchronous operations for a wide range of drivers including GPIO, SPI, I2C, TIMER, PWM, and I2S. While it provides comprehensive functionality, it's important to note that this is a community-driven effort, meaning it might lag behind the latest stable ESP-IDF versions and requires more documentation. For officially supported HALs, users might consider esp-hal, which is no_std-only and requires async programming.

Primis AI, a generative AI solution for hardware design, particularly for FPGA engineers, has officially rebranded to ChipNexus. The platform, which previously offered tools like RapidGPT for streamlining the design process through a natural language interface, now operates under its new identity. The core mission remains to enhance productivity and accelerate the time-to-market for hardware engineering projects, guiding engineers from the initial concept stages through to bitstream generation. All previous functionalities and services associated with Primis AI are now accessible via the ChipNexus brand, with automatic redirection ensuring a seamless transition for existing and new users.

Tensorflow_Object_Tracking_Video is an open-source project developed for object tracking in videos, encompassing localization, detection, and classification. Originally created for the ImageNET VID competition, it leverages TensorFlow technology. The project integrates popular object detection systems like YOLO (You Only Look Once) and TensorBox, along with Inception for classification. It features a modular architecture that includes a general object detector, a tracker, and a smoother. The repository provides scripts for both YOLO and VID TENSORBOX usage, allowing users to process videos, set parameters, and obtain real-time object tracking results. It also includes dataset scripts for preparing and processing data for training, particularly for the VID classes, and offers pre-trained weights for Inception and TensorBox.

This GitHub repository offers a comprehensive tutorial for training a TensorFlow object detection classifier to detect multiple objects on Windows 10, 8, or 7. It covers the entire process, from installing necessary software like Anaconda, CUDA, and cuDNN, to setting up the TensorFlow Object Detection API directory structure. The tutorial details how to gather and label pictures, generate training data, create a label map, configure and train the model, and finally, export and test the inference graph. It also includes Python scripts for testing the classifier on images, videos, or webcam feeds, and provides files for training a "Pinochle Deck" playing card detector as an example.

v8js is a PHP extension designed to embed the Google V8 JavaScript Engine directly into PHP applications. This powerful integration enables PHP developers to execute JavaScript code within a secure sandbox environment. Key features include the ability to set time and memory limits, providing a robust mechanism for safely running untrusted JavaScript code. It supports PHP 8.0.0+ and V8 9.0 or higher, ensuring compatibility with modern development stacks. The extension also offers comprehensive APIs for managing JavaScript execution, handling module loading, and converting data types between PHP and JavaScript, making it a versatile tool for extending PHP functionality with JavaScript capabilities.

visual_anagrams is an open-source tool specifically designed for generating multi-view optical illusions. It leverages advanced diffusion models to create these unique visual effects. The tool offers readily available code, making it accessible for hands-on experimentation. It also includes Colab notebooks, catering to both free and Pro tier users, to facilitate the creation of visual anagrams and exploration of factorized diffusion techniques. This makes it a valuable resource for those interested in the intersection of AI and visual art.

Voicetree is an open-source spatial Integrated Development Environment (IDE) specifically built for orchestrating multiple AI agents. It features an interactive graph-view interface, enabling users to work directly within a visual representation of their AI agent ecosystem. Within this environment, nodes can serve various purposes, including representing markdown notes or acting as terminal-based AI agents such as Claude Code and Gemini. A key capability of Voicetree is that agents can spawn sub-agents and access nearby nodes to gather context, facilitating complex AI workflows and interactions.

yolov5_obb is an open-source project that extends the popular Yolov5 framework for oriented object detection. It integrates Circular Smooth Label (CSL) to accurately detect objects with arbitrary rotations, making it highly suitable for specialized computer vision tasks. The repository provides pre-trained models and detailed results on DOTA datasets, including mAP scores for various versions and speed benchmarks on different hardware. Users can reproduce examples for validation and testing, and the project includes comprehensive documentation for installation and getting started. It's a valuable resource for researchers and developers working on rotation detection in aerial imagery and similar domains.

Zep is an embeddable editor designed for developers, featuring optional Vim keystrokes and a rendering-agnostic architecture. It can be integrated as a shared modern-cmake library or a static library, with a core that is dependency-free. Zep supports drawing to Qt Widgets or ImGui windows, making it suitable for embedding in game engines or other applications. Key features include a simple syntax highlighting engine, basic theming, window tabs, and vertical/horizontal splits. While heavily influenced by Vim, it also offers a notepad-style editing mode. Zep includes a search feature (Ctrl+P) and a Repl mode for console integration, making it versatile for live-coding environments. It is built with a minimal mode and configurable options via a toml-format file, and is cross-platform, supporting Windows, Linux, and Mac OS.

gpt2-ml is an open-source implementation of the GPT-2 model, designed to support multiple languages. It comes equipped with pre-trained models, allowing users to get started quickly, and offers simplified training scripts to facilitate customization. This tool is particularly useful for developers and researchers who want to experiment with and adapt GPT-2 for various natural language processing applications. Its multilingual capabilities make it suitable for projects that require text generation or understanding across different languages.

Hound is an open-source, language-agnostic AI auditor designed for deep and iterative code reasoning. It autonomously builds and refines adaptive knowledge graphs to model various aspects of a system, such as architecture, access control, and value flows. Key features include graph-driven analysis, relational graph views for cross-aspect reasoning, and a belief and hypothesis system that evolves with confidence scores. Hound employs dynamic model switching, using lightweight 'scout' models for exploration and heavyweight 'strategist' models for deep reasoning, optimizing both cost and efficiency. It supports strategic audit planning, balancing broad code coverage with focused investigation, and is optimized for small-to-medium sized projects like smart contract applications.

aio-usb-drive is an open-source project offering a curated collection of diagnostic and rescue tools, operating systems, and application installers, designed for creating a multiboot USB drive. This repository serves as a comprehensive reference for building a "Swiss-army knife" USB drive that bundles essential utilities into a single, updatable, and portable toolkit. It includes a step-by-step guide for preparing and using the USB drive, covering prerequisites, setup with Ventoy, and adding programs. The collection features both open-source tools like SystemRescue, Rescuezilla, and various Linux distributions, as well as closed-source options such as Hiren's BootCD PE and Windows 11 installers. The project aims to simplify system administration and recovery tasks by providing a convenient way to carry and deploy various utilities from one USB drive.