Project Library

An AI-based method that can convert user sketches into refined images. Users can draw a sketch and then use the tool to convert it into a high-quality image. The goal of this project is to provide a convenient way for users to easily turn their ideas into beautiful images.

This project provides a proxy for users of the MidJourney Discord channel, allowing them to call AI drawing functions through API calls. Such a proxy can simplify user interaction with AI drawing functions, making the calling process more convenient. This project belongs to the fields of image processing and artificial intelligence, providing users with a fast way to realize AI drawing.

A project for generating 3D texture shapes from images. Its main goal is to generate high-quality 3D texture shapes from existing image data, which can be applied in the fields of computer graphics, computer vision and artificial intelligence.

🚀🎬ShortGPT - An experimental AI framework for automatic short/video content creation. It enables creators to quickly produce, manage, and deliver content using artificial intelligence and automation.

This is a powerful tool that uses multiple models to increase the image super-resolution to any size, thereby improving the resolution and quality of the image, making it clearer and more detailed. Currently, it supports multiple super-resolution models, including RealCUGAN, RealESRGAN, Waifu2x, and SRMD.

It uses a single model trained across multiple attribute domains to generate facial features and expressions, mainly including facial attributes and expression manipulation including modifying facial attributes such as hair or skin color, age, gender and facial expressions, making them happy, sad or angry.

Ecoute is a real-time transcription tool that provides real-time transcription for both the user's microphone input and speaker output in a text box. This means it can record your speech and convert it into text, as well as capture the speaker's voice and perform real-time text transcription. In addition, Ecoute integrates OpenAI's GPT-3.5, which not only provides real-time transcription but also generates real-time transcription suggestions based on the conversation, offering more intelligent and responsive features.

This project is a flexible interactive video object tracking and segmentation tool based on technologies such as Segment Anything, XMem, and E2FGVI. It provides a convenient way to help users track and segment objects of interest in videos, providing practical tools for video analysis and processing.

Facecchain is a deep learning tool chain for generating your digital twin.

A powerful AI image editor that can create and transform images with simple commands to help users realize creative image editing.

Convert static pictures to dynamic pictures, suitable for short video scenarios

A next-generation face changer and image enhancer. It uses advanced image processing technology, allowing users to blend different facial features together to create fun and impressive effects. The potential applications of this project include entertainment, virtual makeup, and artistic creation, providing users with creative tools.

A machine learning algorithm implementation library open-sourced by Twitter, which includes many commonly used machine learning algorithms and models covering tasks such as classification, regression, clustering, etc. This project provides a convenient resource for machine learning practitioners to quickly acquire and use various algorithms to solve practical problems.

This creative animation tool uses object detection models, pose estimation models and image processing-based segmentation methods to quickly create digital versions of drawings and deform them through traditional computer graphics techniques to make animations.

This project provides the latest technology for real-time object detectors. It is an advanced tool that can be used for object detection tasks in real-time scenarios with efficient and accurate performance.