VIDEO
HIGHLIGHT
Log InSign up
Create AI MOVIES with this FREE tool! Full Tutorial + Project Files
SummaryTranscriptChat

Create AI MOVIES with this FREE tool! Full Tutorial + Project Files

AI Video Generators and the Horrors of AI Documentary Creation

In this section, the speaker discusses the progress made by AI video generators and their use in creating a nature documentary. However, unexpected challenges arise during the process.

Combining AI Tools for Documentary Creation

  • The speaker combines multiple new AI tools to create a BBC planet Earth style nature documentary.
  • The goal is to create unique and never-before-seen animals using AI video tools.
  • The speaker plans to ask an AI model for appropriate names and write short Wikipedia articles about these creatures.

Impressive AI Video Generator - Animate Diff

  • Animate Diff is a framework for animating text-to-image models.
  • It can be installed locally on a computer without requiring monthly subscriptions or complex interfaces.
  • Despite initial hesitation, the speaker finds that Animate Diff works well even on older graphics cards.

Using Com UI for Workflow Building

  • Com UI is a note-based interface for stable diffusion.
  • Each node represents a specific task, allowing users to build custom workflows quickly.
  • Installation of Com UI is straightforward, with tutorials available in the video description or on Patreon.

Generating Octopus Crab Animation

  • The speaker provides free workflows for short and long videos using Com UI.
  • Dream Shaper is chosen as the model for generating animation.
  • Scene dimensions are set to 480p with a 16:9 aspect ratio.
  • Prompts are added to guide the generation process.

Enhancing Video Quality

  • Frame interpolation is used to increase the frame rate from 8 frames per second (fps) to 24 fps.
  • A custom interpolation node is added through Com UI manager.
  • The multiplier is set to three for better smoothness in motion.
  • Output settings are adjusted accordingly.

Impressive Results and Conclusion

The speaker concludes the video by highlighting the impressive quality of the generated animation and expressing satisfaction with the outcome.

Final Video Quality

  • After just four minutes, the final video is ready.
  • The movements and details in the animation are praised as superior to previous methods.

Conclusion

  • The combination of AI tools, such as Animate Diff and Com UI, allows for creative and high-quality video generation.
  • Users can experiment with different models, prompts, and settings to achieve desired results.

This summary provides an overview of the main topics discussed in the transcript. For a more detailed understanding, please refer to the original transcript.

New Section

In this section, the speaker explains each step of their workflow, walks through the com UI installation process, installs custom nodes, prompts the final video, and provides additional tips and tricks for creating videos.

Workflow Explanation

  • The speaker explains each step of their workflow in detail.
  • They walk through the com UI installation process.
  • They install custom nodes to enhance their video creation process.
  • The speaker prompts the final video by providing instructions and guidance.
  • Additional tips and tricks are shared to help create high-quality videos.

New Section

In this section, the speaker discusses their experience with using C GPT to generate AI-generated images and content.

Using C GPT for Image Generation

  • The speaker shares their excitement about discovering a unique creature in a video and decides to use C GPT to identify it.
  • Despite initially fooling C GPT with an AI-generated image, they proceed to ask for a fitting name and a short Wikipedia-style article about the creature.
  • C GPT successfully understands the creature and provides a suitable name (Luminara Octo Glide) along with a description.

New Section

In this section, the speaker continues using C GPT to generate content for other animals and explores voice generation options.

Generating Content for Other Animals

  • The speaker repeats the same process of using C GPT to generate content for other animals like "The Hoppah Hoot," which is a mix between an o and a frog.
  • They ask C GPT to write a planet Earth-style nature documentary about these creatures.
  • The generated content exceeds expectations by accurately describing the creatures' characteristics.

Voice Generation Options

  • To find a suitable voiceover artist for their video project, the speaker turns to 11 Labs, a platform for generating voices.
  • They search through custom voices and find a voice similar to David Attenborough's named George.
  • The speaker mentions playing around with sliders to generate multiple versions of the same text for editing purposes.

New Section

In this section, the speaker explains the final steps of their video creation process, including editing software and effects.

Finalizing the Video

  • The speaker uses DaVinci Resolve, a free editing software, to put everything together and add music and sound design.
  • They also utilize DaVinci Resolve for color grading and adding additional effects like smudges on the lens, camera shake, rim or blur effects.

New Section

In this section, the speaker introduces an AI-generated nature documentary about creatures in a fictional world called Enchantis.

Enchantis: A Sun-Drenched Landscape

  • The narrator sets the stage by describing Enchantis as a sun-drenched landscape where life thrives in unexpected forms.
  • Two creatures are introduced: the Sand Scuttle and the Hopper Hoot.
  • The Sand Scuttle makes its way across the sands while being observed by the Hopper Hoot.

New Section

In this section, more creatures from Enchantis are introduced along with their unique characteristics.

Luminara Octo Glide: A Radiant Creature

  • The Luminara Octo Glide is described as a creature of luminescence and continuous metamorphosis.
  • It feeds on bioluminescent algae, which adds to its radiant allure.

Sephila Shade: A Lurking Nightmare

  • The Sephila Shade is introduced as a creature of Enchantis' darkest depths with insatiable hunger.
  • It is driven by a parasitic organism within, leading it to hunt incessantly.
  • The Sephila Shade's pursuit of the Luminara Octo Glides creates a fascinating spectacle.

New Section

In this section, the narrator describes the captivating behavior of the Luminara Octo Glides and their defense mechanism against the Sephila Shade.

Radiant Defense

  • The Luminara Octo Glides gather together, creating a blinding light through their combined luminescence.
  • This radiant display serves as both a mesmerizing spectacle and a strategic defense mechanism.
  • The Sephila Shade, unaccustomed to such brilliance, recoils in disorientation.

New Section

In this section, the narrator concludes the nature documentary about Enchantis and its creatures.

A Mesmerizing Tapestry

  • As night falls and darkness envelops Enchantis, the Luminara Octo Glides gather to celebrate their victory over the Sephila Shade.
  • Their glowing presence transforms Enchantis' shores and waters into a mesmerizing tapestry of light.
Video description

I’m showing you how I used a NEW AI TOOL to create a terrifyingly good nature documentary! If you like my work, please consider supporting me on Patreon: https://www.patreon.com/Mickmumpitz And follow me on Twitter/X: https://twitter.com/mickmumpitz There are a lot of new AI video generators like Runways Gen-2, Pika Labs or moonvalley. But the most impressive one for me is AnimateDiff. It’s a framework for animating txt to image models and you can install it locally on your computer. So no weird Discord interfaces, or costly subscriptions are necessary! 👉 You can get my FREE WORKFLOWS here: https://www.patreon.com/posts/free-workflows-92142920?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link TOOLS USED IN THIS VIDEO: - ComfyUI https://github.com/comfyanonymous/ComfyUI - AnimateDiff https://github.com/Kosinkadink/ComfyUI-AnimateDiff-Evolved - Elevenlabs https://elevenlabs.io/ - DaVinci Resolve https://www.blackmagicdesign.com/de/products/davinciresolve TOOLS SHOWN IN THIS VIDEO: - runway GEN - 2 https://research.runwayml.com/gen2 - PikaLabs https://www.pika.art/ - Moonvalley https://moonvalley.ai/ RESOURCES + TUTORIALS: - ComfyUI Installation: https://github.com/comfyanonymous/ComfyUI - Animate Diff Installation: https://github.com/Kosinkadink/ComfyUI-AnimateDiff-Evolved - Guide on which all my workflows are based: https://civitai.com/articles/2379/guide-comfyui-animatediff-guideworkflows-including-prompt-scheduling-an-inner-reflections-guide - Great beginner tutorial: https://youtu.be/kmZ5S2X55fU?si=joYWTlDfZxH5rnVF CHAPTERS 00:00 Intro 00:30 Concept 01:20 The Tool 02:47 ComfyUI Settings 06:04 Script 07:06 Voice + Effects 07:55 FINAL FILM! Mickmumpitz is a channel that is all about understanding new technologies from the worlds of AI, digital art, film and games in a playful and project-oriented way.