Midjourney + Runway Gen 3 Turbo Full Course (Create Lifelike Cinematic AI Videos)

The script introduces the era of AI films and the potential democratization of filmmaking with Runway Gen 3 Turbo. It highlights the advancements from the initial use of Gen 2 to the current capabilities, which are nearing super photorealism and coherence. The speaker shares their experience in pushing the boundaries with AI tools and provides a tutorial on creating a mini AI film. The video is structured into three parts: storyboarding, transforming storyboards into cinematic shots with M Journey, and creating cinematic videos with Runway's image-to-video mode. The process involves testing various ideas and sharing practical tips based on the official Runway Genry prompting guide. The workflow begins with writing a synopsis, choosing a character and setting, and then using CET GPT cloud or similar models to generate scenes. The speaker emphasizes the importance of specifying setting, location, and time for consistency in image generation.

This paragraph details the process of upgrading a storyboard from low to high fidelity using Mid Journey's image prompting and style reference. The speaker explains how to use image prompting alongside detailed visual text descriptions from CAD GPT to enhance the image conversion to video. They discuss the importance of image weight in balancing the image and text prompts and share reference values for convenience. The paragraph also covers the use of previously generated images as style references and provides tips for refining character depictions and directing upcoming scenes. The speaker demonstrates how to use character weight to maintain consistency in character appearance across different scenes and how to combine image prompts with shot types to enhance the image-to-video process.

The speaker discusses the creation of a cinematic universe using image prompting, focusing on the use of brackets for permutations or batch prompting in Mid Journey. This technique allows for multiple image generations with a single prompt, with the 'chaos' parameter controlling the variation among the generated images. The speaker recommends staying within a chaos value range of 4 to 8 for maintaining prompt adherence while allowing for variety. They also mention the use of style references to ensure consistency with the rest of the generated images. The paragraph concludes with tips for handing over the generated images to Runway for further processing, including organizing images through liking and folder utilization, and the importance of upscaling for quality enhancement.

This paragraph delves into the use of Runway's image-to-video capabilities, emphasizing the speed and quality of Genry Turbo. The speaker describes how to set camera movements based on the first frame and the creative direction, whether starting from or leading to a specific frame. They mention the addition of prompt presets for easier camera control and the official prompting guide for text-to-video and image-to-video. The guide suggests focusing on camera movement, character, and scene movement rather than describing the image itself. The speaker provides examples of how to structure prompts with camera techniques, follows X format, and events, and how to combine these elements for cinematic footage. They also touch on the use of keywords for speed control and maintaining color consistency.

The speaker explores advanced prompting techniques for image-to-video, emphasizing the simplicity of using the 'camera follows X' format and the importance of describing camera movement. They provide examples of how to combine camera techniques, movement, and events in a single prompt and discuss the flexibility in prompt construction based on the image and desired complexity. The paragraph also addresses common problems with Runway, such as slow motion and color changes, offering solutions like using speed-related keywords and adjusting color consistency with specific prompts. The speaker highlights Runway's ability to depict emotions and demonstrates the lip sync feature, which allows for character speech and hand gestures based on user input. They conclude with a discussion on Runway's pricing, suggesting a low-resolution preview option for better cost control and user satisfaction.

In the final paragraph, the speaker introduces the concept of a fixed seed on Runway, which creates visually similar generations akin to style references in Mid Journey. They explain that the fixed seed is unchecked by default but can be used to maintain a consistent style and moment in the output. The speaker concludes the video by encouraging viewers to support the content and join the community for more tutorials on cinematic AI video making.