Deonis Kiroshka

699 posts

Deonis Kiroshka banner
Deonis Kiroshka

Deonis Kiroshka

@deogenai

Exploring realism in generative AI. Focused on creating lifelike visuals that capture candid moments.

Katılım Eylül 2025
79 Takip Edilen209 Takipçiler
Sabitlenmiş Tweet
Deonis Kiroshka
Deonis Kiroshka@deogenai·
A short story "Sail Anyway". A reflection of the thoughts that've been on my mind lately. Let me know what you think.
English
11
2
32
4.4K
Jordan Daniel Chesney
Jordan Daniel Chesney@jordandchesney·
In case you haven't noticed, @invideoOfficial values AI filmmakers more than anyone in the game right now. I just had a chance to hang with them at @aionthelot, and it was such a please. Big hearts, big smiles, and not a shred of pretension that's part and parcel for most Hollywood types. Just genuine guys that genuinely care about storytellers. @invideoOfficial, here are a bunch on great AI Filmmakers I strongly recommend you engage. I'd say trust me, but all you have to do is take one look at their work and you'll see what I see: @ai_artworkgen @rmuir3 @DavidLaChanceJr @TheBLUDSIMPLE @alexgnewmedia @h_khachatrian @mind_wank @piotrbinkowski @BengtTibert @NathanBoey @deogenai @jdp2oo @JossMonzoni @Julianc_AI @julienvallee @NomadsVagabonds @Ahn_Jae_Hong @BLVCKLIGHTai @piotrbinkowski @olgabrnv @cfryant @henrydaubrez @jdp2oo @drdreams @elena_sav_ @gen_ericai @gabemichael_ai @rourke_heath @Jett_Collective @panaviscope @tupacabra @UnoParticular @eattheethos
English
12
4
52
2K
Dr. Dreams
Dr. Dreams@DrDreams·
AI videos are popping off today! This is a proof of concept that was financed by Steven Spielberg and Amazon Creator Studios. @Grok please get this video in front of nerds, wizards, cyber hackers, body builders, long haul truckers and marc andreessen. Would also accept naval
English
35
18
165
18.6K
panaviscope
panaviscope@panaviscope·
1848 - Trailer.
English
12
10
41
3.8K
Christopher Fryant
Christopher Fryant@cfryant·
Now we're talking! Elevenlabs new music model can get super experimental. Up to and including making a techno song made by and for mosquitos. Unironically, this is what I've been wanting for a long time.
Johan Nordberg@almost_digital

🦟

English
10
2
40
3.3K
Deonis Kiroshka
Deonis Kiroshka@deogenai·
@cfryant clients ask about that. It would be really cool to be able to download the sounds.
English
0
0
1
32
Christopher Fryant
Christopher Fryant@cfryant·
One thing I'd like to see from next gen AI video models is the ability to download audio stems. -Voice track -Individual sound effects -Music (with their own stems if possible) This is often vital to really push a good segment to something pro level. What else do we need?
English
22
2
60
2.3K
fofr
fofr@fofrAI·
Gemini Omni Flash: > a recording from a capsule on the london eye, a jerky zoom into something in the distance and then refocusing (with a bit of back and forth) (no timestamp or dialog) Note the world knowledge of London’s landscape, and the way the video is gently moving like the capsules do.
English
25
30
491
231.3K
Google DeepMind
Google DeepMind@GoogleDeepMind·
We’re dropping Gemini Omni: our first step towards a model that can create anything from anything - starting with video. It combines Gemini’s intelligence with our generative media systems - representing a leap forward in world understanding, multimodality, and editing 🧵
English
412
1.3K
8.4K
1.2M
Jordan Daniel Chesney
Jordan Daniel Chesney@jordandchesney·
@deogenai I think we have to be relentless. If we relentlessly work on the stories that are close to our hearts then we will naturally form a signature and style that is ours and ours alone. Then, and possibly only then, will others view us as storytellers instead of button pushers.
English
1
0
2
55
Henry Daubrez 🌸💀
Henry Daubrez 🌸💀@henrydaubrez·
A lot of the opportunities I got from traditional studios and industry players over the last few months were essentially to be “the AI guy.” Not directing. Not writing. Not building stories. Just the person who knows how to operate the tools. A technical role. The guy in the basement who knows how the machine works. And I’m definitely not the only one seeing this. Which is also why I think this is the right moment to be careful about getting locked into being only the tooling person internally. There’s nothing wrong with those opportunities, and for some people they’re genuinely great paths, but it’s worth remembering how fast this space is moving. The workflows will evolve. The tools will level out. Eventually most people will know how to use them. The most unique thing about you is still you. Your taste, your ideas, your storytelling, your point of view. Not your workflow.
Henry Daubrez 🌸💀 tweet media
English
27
5
163
6.5K
Jordan Daniel Chesney
Jordan Daniel Chesney@jordandchesney·
No more need for actors means you just prompt "good performances", right? WRONG. Good performances are simply choices. Maybe the actor thinks they gave a great performance, but if the director thinks otherwise, then you go again. Strong directors simply have the instinct to discern a good performance, AI or real, it doesn't matter. A real actor can feel fake, and an AI performance can feel real. Eventually, mass audiences won't care what's an AI performance and what's real because all of cinema will be saturated with the technology, and they simply won't be able to tell. If anything, they'll just grow tired of guessing and accept it. But none of this tech changes the fundamentals... a good performance is a decision, and now both the creation of the performance and the conclusion that it's a good take is down to the director... And Deo (@deogenai) is an absolute natural born director.
Deonis Kiroshka@deogenai

A short story "Sail Anyway". A reflection of the thoughts that've been on my mind lately. Let me know what you think.

English
3
0
26
1.6K
gen.ericai
gen.ericai@gen_ericai·
Mundos paralelos
Português
1
0
11
458
Kōda
Kōda@aimikoda·
A new experiment with Seedance 2.0 on @mitte_ai : FACS. For this videos, I only used FACS codes in the prompt. I didn’t describe the facial expressions in plain language at all. FACS (Facial Action Coding System) is a system for describing facial expressions using individual muscle movements called Action Units (AUs), instead of general emotion labels like “happy” or “sad.” It breaks the face into controllable components such as brow movement, eyelid tension, lip movement and cheek activation. Even though it didn’t follow all 14 Action Units perfectly, it still interpreted most of them surprisingly well. I think this could become especially useful for subtle facial acting, forced smiles, uncanny expressions, mixed emotions and micro-performance details during close-up dialogue shots. You can check the prompt for first 15s, a list of FACS codes generated with GPT Image 2 in the replies.
English
63
99
1.1K
442.9K
Deonis Kiroshka
Deonis Kiroshka@deogenai·
Recently I took part in an AI competition by MyFilm48 where the challenge was to make a movie trailer in 96 hours. I've never made a trailer before, so it was an interesting experience. Let me know what you think.
English
5
0
16
812
MayorkingAI
MayorkingAI@MayorKingAI·
Share your AI Art Have you made something amazing with AI? Images, videos, anything goes! Tomorrow (Sunday), I’ll feature the top creations in a special post Drop your best work below 👇
English
143
9
114
7.2K
Marco Magario
Marco Magario@MarcoMagarioAI·
HOLY GRAiL — a comedy set in Renaissance Florence, 1503. Written, created, and scored by me in ten days. Entirely AI-generated. This film got me noticed by @thedorbrothers. I have the honour to work with them now. It’s gonna be a crazy journey. Full film on YouTube: @MarcoMagarioAI" target="_blank" rel="nofollow noopener">youtube.com/@MarcoMagarioAI
English
99
62
546
80.1K