Total reading time around 4 minutes.
Welcome to Visually AI!
🎉Midjourney Year One
I am thrilled and honored to be included in Midjourney’s Year One archival book to celebrate Midjourney’s development and the community’s exploration over the past year.
You can purchase a copy of the book or previous issues of Midjourney Magazine here.
I was surprised and overwhelmed to see my name trending on 𝕏:
This accomplishment means a lot to me, because I never thought I would be able to master Midjourney when I started using it in December, 2022.
I don’t have a background in Art or Design, but I am a creative person who loves learning new skills.
I encourage you to pursue your passions, no matter where you are in your life ❤️
🔮AI News This Week
🎬Gen-3 is here for everyone!
Runway has launched Gen-3 Alpha, its latest AI video generation model, making it available to all paid users on the platform.
The quality is incredible, especially since it only takes text prompts right now. I believe the results will improve even more when we can use images as prompts.
Sora has major competition and it is still not available…
Gen-3 Key features:
Generates 5-10 second high-quality videos from text prompts:
Prompt: Low-angle tracking shot following a sleek sports car. Lighting: Neon lights reflecting off wet pavement, creating a vibrant and dynamic scene.
Enables creation of dynamic text and title sequences:
Prompt: Blockbuster cinematic title animation of Northern lights forming the word "AURORA" in the twinkling night sky.
Allows for fine-grained control over camera movements and scene transitions:
Prompt: Rotating shot: A skilled chef prepares a gourmet dish in a high-end kitchen, chopping ingredients and plating with precision
Ability to generate people and faces with detailed skin textures and expressions:
Prompt: Close-up: An elderly man's face transitions from sorrow to joy as he reunites with a long-lost friend. Soft, warm lighting enhances the emotional impact.
How to prompt for Gen-3
To effectively prompt Gen-3, start with a clear description of the camera movement, followed by the main scene details. Then, add any additional specifics about the action, subjects, or style you want to see.
Runway has a Gen-3 prompting guide with visual examples.
The recommended format is:
[camera movement]: [establishing scene]. [additional details].
I used that format for the clip below:
Prompt: Timelapse: A barren desert blooms as sand transforms into a vibrant wildflower field. Clouds race overhead, casting moving shadows.
Common elements for effective Gen-3 prompts:
Camera Movement: Specific description of how the camera moves or behaves.
Example: "Slow tracking shot moving from left to right"
Scene Description: Clear establishment of the setting and main subject.
Example: "A bustling Tokyo street at night, focusing on a neon-lit ramen shop"
Action or Change: Description of what happens or changes in the scene.
Example: "A chef expertly prepares a bowl of ramen, steam rising from the broth"
Visual Details: Specific visual elements, textures, colors, or lighting.
Example: "Soft orange lantern light illuminates the chef's weathered hands"
Atmosphere or Mood: Words that convey the overall feel or tone.
Example: "A cozy, inviting atmosphere with a sense of culinary tradition"
I’m working on a longer tutorial and guide. I’ll let you know as soon as it’s available!
You could have your AI service, tool, or event seen by Visually AI’s community of over 15,000 subscribers:
🚀 This Week’s AI Tools
gpt4All 3.0 Desktop app: Run LLMs locally on your desktop - no internet required. (link)
Promethean: Digital asset management; AI assistant designed to significantly accelerate digital art production for creative teams. (link)
Moshi: AI voice chat platform for natural, conversational interactions with advanced NLP and speech recognition. (link)
ElevenLabs Voice Isolator: free AI-powered tool that effectively removes background noise from audio recordings, enhancing speech clarity without distortion. (link)
Rapport: Create and deploy interactive, AI-driven virtual personalities for engaging conversations across various platforms and devices. (link)
💻Visually AI on YouTube
I posted a showcase of my first results using Gen-3 Alpha, with prompts:
Runway Gen-3 Results - What can it produce? Prompts included (No Editing)
🖼️ Image Prompts
Prompt: A cinematic still of a man crouching on a rooftop at dusk, high angle shot, warm desaturated tones with heavy shadows, intense focus on the character's determined expression, shot on an Arri Alexa, 8K
Prompt: A chameleon walking across a canvas, its skin changing to mimic famous paintings with each step, leaving behind a trail of blended art styles.
Thanks for reading, and have a creative week!
The combination of these AI tools now fully enables the creation of textured videos. I also often use Midjourney. Midjourney excels at imitating specific artistic styles.