Any pose in MJ: ECU on a detail and then ZOOM OUT

Glibatree (Ben Schade) recently implored on YouTubeDo THIS to Create Amazing Poses in Midjourney!!!

The problem with a lot of image generators is that they love selfies: front-facing portraits. But what if you want a profile? Ben has a two-step work-around:

“Generate a close-up photo of your subject’s ear and then use the editor to zoom out and create the rest of the image.”

He explains:

“The reason this works is because what Midjourney needed was a pattern interrupt. Take advantage of its usual way to generate images by finding the usual way to generate an image with a more unusual focus. It’s better to choose a focus that is already often viewed from the angle we want.

  • focus on a ponytail if we want to see the back of someone’s head
  • use a receding hairline to see someone from straight above
  • focus on the back pocket of a pair of jeans if you want the…
  • I wouldn’t recommend looking up someone’s nostril (I mean it’s an angle that works but I just wouldn’t recommend it.)

The point is we can generate any of these things using extremely simple prompts and get very unusual angles to be seeing a person from. And then starting from there once we have the angle well defined we can simply zoom out and make our chosen feature less prominent by changing our prompt to something else and so in the new image the angle we wanted is extremely well defined not by tons of keywords but by the part of the image we already generated.”

This works for Expressions as well. He explains:

“If we start with a photo of just a smile or just closed eyes or just a mischievous smirk, Midjourney will spend all of its effort to create a high quality closeup version of the exact expression we wanted that now, in just one more generation, we can apply to our character by simply zooming out.”

My take: thank you, Ben, for cracking the code!

Be the hero of your own story.

Jason Hellerman reminds us on No Film School that No One Is Coming To Save Your Filmmaking Career.

He confesses:

“The hardest Hollywood truth I have had to come to terms with is that no one is going to come and save me.”

“I’m going to have to save myself.”

“I talk to so many young writers and directors who think someone needs to pluck them from obscurity so they can begin their careers. That’s just not true. Nothing is holding you back except your willingness to work hard and to create new things to show people. Start with small budgets, write things you can shoot. Then build from there. Get really good, get undeniable. Sure, to have a career in Hollywood a little luck is involved, too. But I am a firm believer in making your own luck.”

  • Keep learning.
  • Keep creating new things.
  • Keep putting yourself out there.
  • Be your own advocate and your own voice.
  • Stay relevant.

He concludes with, “It is on you. You’re the hero of your story.”

My take: I really needed to be reminded of this, this week. Thanks, Jason.

FaceFusion 3: the best free face swapper

Tim of Theoretically Media has a great review of FaceFusion 3.0.0 on YouTube:

In it he discusses:

  1. How to install FaceFusion 3 using Pinokio
  2. How to face swap for video
  3. The limitations of FaceFusion
  4. Face swapping with AI-generated characters
  5. Lipsync
  6. Expression controls
  7. Aging controls

A huge bonus to this pipeline is face_editor. See 14:02 for tools to alter the many elements on faces, such as smiles, frowns and eye lines. Even age.

My take: we are way beyond deep fakes now. The ability to change expression is extremely powerful! Every performance can be altered.

Flux.1 prompting and guidance guides

CyberJungle, the Youtube channel of Hamburg-based Senior IT Product Manager Cihan Unur, recently posted a great video on consistent generated characters.

There are lots of great insights in this 20-minute video. Two outstanding takeaways:

First: a prompting guide for Flux.1. At 15:28 he reveals three prompting styles: list, natural language and hybrid.

Second: a guidance guide for Flux.1. At 17:18 he shows Photorealistic and Cinematic images with a wide scope of guidance values. He posits:

“The essence of guidance setting is a compromise or a balance between photo realism and prompt understanding.”

See 18:36 for the Photorealistic results. He prefers a level of two.

See 19:54 for the Cinematic guidance level he prefers: again two.

My take: to me, too often generated images look over-the-top and so ideal, they’re unrealistic. The key seems to be dialing the guidance down to two. Who knew? Now, you do.

New Generated Video pipeline?

A couple of very recent videos point to a potential new Generated Video, or GV, pipeline.

The first is “Create Cinematic Ai Videos with Kling Ai! – Ultra Realistic Results” by Seattle’s Yutao Han, aka Tao Prompts.

The second is “How-To Create Uncensored Images Of Anyone (Free)” by Lisbon’s Igor Pogany, aka The AI Advantage.

Imagine combining both into a new GV pipeline:

  1. Train custom character models
  2. Create key frames utilizing these custom models
  3. Animate clips with these key frames
  4. Upscale these clips
  5. Edit together.

My take: a lot of people will immediately claim this is heresy, and threatens the very foundations of cinema as we’ve come to know it over the last one hundred years. And they would be right. And yet, time marches on. I believe some variation of this is the future of ultra-low budget production. Very soon the quality will surpass the shoddy CGI that many multi-million dollar Hollywood productions have been foisting on us lately.

Compare Image Generators at a glance

Matt Wolfe has just released a wonderful comparison of top image generators tackling four different types of pictures on YouTube.

The four image categories are:

  • Human Realism
  • Landscapes
  • Scenery incorporating Text
  • Surrealistic Images

The platforms are:

  1. Ideogram 2.0
  2. MidJourney 6.1
  3. Mystic
  4. Phoenix
  5. Flux.1 (Grok)
  6. Dall-e 3
  7. SD3
  8. Firefly 3
  9. Meta Emu
  10. Imagen 3
  11. Playground v3

See the Figma board to see all eleven contenders at once.

My take: as a visual learner, I really appreciate this side-by-side comparison. Thank you, Matt!

August 2024 AI Video Pipeline

Love it or hate it, as of August 2024, AI Video still has a long way to go.

In this video, AI Samson lays out the current AI Video Pipeline. Although there are a few fledgling story-building tools in development, full-featured “story mode” is not yet available in AI video generators. The current pipeline is:

  1. Create the first and last frames of your clips
  2. Animate the clips between these frames
  3. Create audio and lip-sync the clips
  4. Upscale the clips
  5. Create music and SFX
  6. Edit everything together offline.

It seems new platforms emerge weekly but AI Samson makes these recommendations:

00:23 AI Art Image Generators
09:19 AI Video Generators
16:28 Voice Generators
18:02 Music Generators
20:44 Lip-Syncing
21:52 Upscaling

Keep an eye open for LTX Studio though.

My take: You know, the current pipeline makes me think of an animation pipeline. It’s eerily similar to the Machinima pipeline I used to create films in the sandbox mode of the video game The Movies over ten years ago:

September 8 deadline for CIFF Pitch Sessions

Folks who follow this blog, know that I love Telefilm‘s Talent to Watch competition. It remains your best chance at funding your first feature film in Canada.

Until they allowed direct submissions from underrepresented folks two years ago, this is normally a two-stage process. Each of approximately 70 industry partners get to forward one (and sometimes two or three) projects to Telefilm and then the Talent to Watch jury selects eighteen or so for funding.

The prize? $250,000. One quarter of a million dollars.

Don’t belong to one of the Industry Partners? No problem!

The Chilliwack Independent Film Festival has got you covered. Launched last year, Pitch Sessions lets you throw your first feature project into the ring; five are selected to then pitch in person at the festival and the winner becomes CIFF’s nominee to Telefilm’s next Talent to Watch competition.

Oh yah, the top five also get free passes and a hotel room for the festival.

The deadline to apply to Pitch Sessions at the 2024 Chilliwack Independent Film Festival is September 8.

My take: If you’ve got a spare $100 and you want to hone your pitch in public, this is a great opportunity. Note that each industry partner sets their own rules but this is the only one I know of that incorporates a live pitch. Just be aware that Telefilm typically doesn’t open the Talent to Watch competition until mid-April.

Deadline approaches for women to apply to BANFF Spark

The application deadline for this year’s BANFF Spark Accelerator for Women in the Business of Media: Producers Edition is rapidly approaching: August 12, 2024.

BANFF Spark provides market access, training, and networking opportunities to help build more Canadian women-owned media businesses.

“Since the program began in 2019, BANFF Spark has already provided opportunities for more than 200 women entrepreneurs. The program is open to all candidates and is designed to empower women of colour, Indigenous women, women with disabilities, 2SLGBTQI+ women, and non-binary individuals.”

All selected participants will receive:

  1. Online workshops (that address the core components of business development).
  2. Networking opportunities with top industry professionals.
  3. A full-access pass to the 2025 Banff World Media Festival (June 8-11, 2025) and its complement of top industry sessions and international marketplace.
  4. A $1500 CAD travel stipend to attend the 2025 Banff World Media Festival
    (on the condition of in-person Festival attendance).

Apply here.

My take: I love that this is focussed on people and not squarely on projects. I don’t have to mansplain this, just apply!

July 2024 Tier List for AI Video

Igor Pogany of The AI Advantage recently released a YouTube video that succinctly summarizes the current state of AI Video.

The tools he reviews are:

His favourites (dated mid-July 2024) are:

Runway GEN-3 Alpha and Luma Dream Machine for their clip outputs, but watch out for LTX Studio because of their overall project approach.

See the full tier list at 12:48 for the tl;dr.

My take: this is a super-valuable video that can get you up-to-date in under 14 minutes. Well worth your time.