Ethan B. Holland

Agenda

• Introductions – around the room
  • Format is informal, I’m not here to present but since I don’t know everyone, I’m prepared with conversation starters and happy to help.
  • Name, where you’re from, a little about yourself
  • What you’re looking to get from tonight’s meeting
  • Familiarity with Runway and any other tools?
  • Traditional experience or other skills, hobbies, that connect your interest to Runway
• Code distribution and stickers
  • Free accounts: Meetup10Milton
  • Paid accounts: TA0YMMAR
  • Each code is worth $10
  • Tutorial on redeeming codes
• Sharing/stories/examples
• Connecting

Openers:

Dr. Eli David on X: “Both videos are AI generated, one in 2023 and the other in 2024. This is how rapid the progress in AI is! (Kling is a Runway competitor)
https://twitter.com/DrEliDavid/status/1800062863811248267

Olivia Moore on X: “Crazy to see how far AI video has come in the last few months The handling of shadows and light here feels like a new standard 👇 (from @endlesstaverns, made with Runway Gen-3 + Suno) https://twitter.com/omooretweets/status/1815113401359880196

Example of building with AI
enigmatic_e on X: “I had to test how Gen-3’s video-to-video would perform using a Viggle output. Have any of you done any interesting with vid2vid? X – https://twitter.com/8bit_e/status/1844435081869525489

Storyboarding and consistency
InterdimensionalTV on X: “WrestleMania of Power Made with: @midjourney @LumaLabsAI @udiomusic Featuring the world’s most powerful individuals at WrestleMania. X – https://twitter.com/n_reruns/status/1822959682518237340

Funny teasing of AI videos
GioM on X: “Some Chinese dudes imitating AI videos lol this is next level https://twitter.com/theGioM/status/1824000938870292771

Examples:

Milton theater Runway meetup video:
https://ethanbholland.com/2024/09/21/behind-the-scenes-runways-video-generation-in-action/

Runway Gen-3 Image-to-Video Testing with MidJourney and Udio: Tech Fashion Show! – Ethan B. Holland – https://ethanbholland.com/2024/07/31/runway-gen-3-image-to-video-testing-with-midjourney-and-udio-tech-fashion-show/

Borriss on X: “This is bananas. Gen-3 Alpha just got dropped by Runway. 7 wild examples: https://twitter.com/Borriss/status/1802731501752045860

Rory Flynn on X: “Runway Gen-3 img-2-vid → Awesome. My day is ruined. https://twitter.com/Ror_Fly/status/1818004461735829831

Runway on X: “Today we are releasing Gen-3 Alpha Image to Video. This update allows you to use any image as the first frame of your video generation, either on its own or with a text prompt for additional guidance. https://twitter.com/runwayml/status/1817963062646722880

Are video models “world models”? Tangentially related to segmentation. Awareness of the objects as it diffuses.

Ethan Mollick on X: “So Runway suspects that any large enough text-to-video system will become a world model. OpenAI has hinted at the same with Sora. Seeing some similar abilities playing with Kling the very impressive text-to-video system from China. If true, this is going to be really interesting
https://twitter.com/emollick/status/1816331031349190672

Runway on X: “Gen-3 Alpha exhibits several simulation capabilities, including the ability to generate dynamic camera motions, complex fluid motion, and interactions between objects. We expect further simulation capabilities to emerge as we continue to scale our models.
https://twitter.com/runwayml/status/1816096185016357030

Runway on X: “3/10 ” / X – https://twitter.com/runwayml/status/1817963066610557103

Runway on X: “8/10” / X – https://twitter.com/runwayml/status/1817963099556823467

Runway on X: “10/10 ” / X – https://twitter.com/runwayml/status/1817963104061505908

nano on X: “Gen-3 Apha is fking insane thread: Prompt: Highly detailed close up of a bacteria. https://twitter.com/nanulled/status/1802714455672869195

Runway on X: “Introducing Gen-3 Alpha: Runway’s new base model for video generation. Gen-3 Alpha can create highly detailed videos with complex scene changes, a wide range of cinematic choices, and detailed art directions. https://twitter.com/runwayml/status/1802691475391566108

Builder:

Flux LoRa: https://fal.ai/models/fal-ai/flux-lora?from_training=7e982915-85f9-4f77-8684-fd18dee6fc09
LoRa trains the AI on faces or style references (for consistency)
Runway: https://app.runwayml.com/video-tools/teams/ethanholland/ai-tools/generative-video

Other related video tools:

What is diffusion?

Two years old video but valuable foundation. Extrapolate this from still images to video… and it illustrates how much computing power goes into Runway.

Mapping/Segmentation

Meyga Vox on X: “@GerdeGotIt Amazing videos! Both the original and the AI generated one work great together👍 X https://twitter.com/meygavox/status/1808121948393607337

Viggle/LivePortrait
Proper on X: “This is LivePortrait – it can animate an image from a video reference with incredible accuracy. More unreal demos and link to the project below 👇 https://t.co/yQ6Qhy5gKl”

Viggle
https://ethanbholland.com/2024/04/08/the-viggle-ai-memes-impact-on-image-to-video-awareness/

enigmatic_e on X: “Luchador Action Figure Animation 💪 Tools I used for this: @ideogram_ai for generating reference images @ViggleAI to transform me into a Luchador @AdobeAE for compositing @ComfyUI to improve the results X – https://twitter.com/8bit_e/status/1828530971164995715

Runway Competitors: Kling

Javi Lopez vacation video
https://twitter.com/javilopen/status/1827077427933122689

Kid eating noodles
https://twitter.com/rowancheung/status/1825911087960641836

Guy eating noodles
https://twitter.com/rowancheung/status/1825911226779463891

Jon Finger on X: “I finally tried @Kling_ai ’s image to video. It only gave credits for 6 tests but I was fairly impressed with how consistent it did what I asked pretty well first try (eg: “clean the debris off the old woman”) I’ll do some creative pipeline oriented tests when I get more credits. https://twitter.com/mrjonfinger/status/1817643812233347317

MadMax Beer Commercial
https://twitter.com/rowancheung/status/1825911155300139045

Kling AI on X: “Kling AI’s drone fly-through effect https://twitter.com/Kling_ai/status/1823275917638283395

PixelWise on X: “Newspaper Mummy Skateboarding on the Street. Aivideo created by kling. #kling #klingai @Kling_ai #midjourney #AIart #aivideo #sport ⛳️img : midjourney 📽️vid: kling 🎼music: capcut 🎬edit: capcut https://t.co/xG7sU5JXKO” / X – https://twitter.com/PixelWiseAI/status/1815224664454742321

Ethan Mollick on X: “Knowing the training data for any AI video maker has a lot of typical YouTube stuff, may I present: A man reacts to rocks, a chef puts far too much salt on some food (nice salt physics), hard math is explained, a woman reacts to a potato, a Twitch streamer, & flaming car review X – https://twitter.com/emollick/status/1809088797784551800

Eclectic Method ♻️ on X: “Kung Fu Cats @LumaLabsAI https://t.co/gbtWukb9qs” / X – https://twitter.com/EclecticMethod/status/1802445254663999588

el.cine on X: “Okay, you have to see this. I asked @Hailuo_AI to show a woman going from happy to sad, then crying and covering her face with hands. I was honestly shocked AF! Over the shoulder shot of a woman’s close https://twitter.com/EHuanglu/status/1833522650846793970

Other Concepts/Demos

Segmentation
https://ethanbholland.com/2024/09/14/trying-metas-segment-anything-2-sam2-demo/

HeyGen
https://ethanbholland.com/2024/05/14/ai-news-category-cover-walkthrough-may-10-2024-part-i/
https://youtu.be/J_FOE5kxUTE?si=gfUgzEPQZW3-SNfL