Midjourney is an up and coming tool that allows users to generate surreal renderings. It especially excels at creating fantasy and dystopian sci-fi scenes with dramatic lighting that look like rendered concept art from a video game. It can also generate images that mimic various art styles. The beta version is operating entirely through Discord. First, users type their prompts into a chat interface. In response, a bot demonstrates their generations rendering in real time. Users can then choose to upscale and enhance an image from each set of generations, or create more variations from the same prompt.
DALL-E 2 is an AI model that can generate images from any text prompt given. The model was trained by analyzing millions of images from the internet with their associated captions. Over time, it learned how to draw an image from a text prompt. Essentially, the process followed is after an image encoder turns raw images into a sequence of numbers with its associated decoder, the model turns a text prompt into an encoded image. The model then judges the quality of the images generated for better filtering.
A.V. Mapping is a one stop AI video/images music matching & licensing platform. Users upload videos or photos to A.V. Mapping through computers, tablets, or cellphones. They can also input the budget, their preference, reference music, or scripts; the AI then analyzes the video and music and provides three suitable songs. Users can then download the preview songs, or reanalyze if they like. SFX and audio editing are ready at the same time. Then, users can check and sign the contract immediately without going back and forth with music or license companies. The company aims to maximize efficiency and productivity for the entire creative industry by providing a one-stop platform of mapping the most suitable music and dealing with the music licensing for every video producer, filmmaker, creative agency, production house, etc., while shortening the film scoring time by 2000% instead of half a year. It also increases the commercial value of every single music note.
2 AI cooperation! When Dall-e 2 (image AI) met with A.V. Mapping (image music matching AI)
Demonstration about A.V. Mapping finding music
Disco Diffusion is an AI modeling technique that is utilized to convert Text-to-Image using prompts that describe a scene. It is a free software, written in Python, that is meant to be run in a Google Colab notebook. Users can create images and videos generated by machine learning technology.