The Cold Call, an AI generated British gangster film. Sound on please! This is a step forward in my workflow to create AI cinema from image2video with expressive, consistent characters. Images are generated in midjourney, animated using 3/4 Runway Gen2 and 1/4 Pikkalabs.…
Do you want your generated images to come alive? Just head over to runway drop your still into the gen2 generator and don’t add a text prompt and then watch the magic happen 24 frames of animation just like that! 🤯 @ me when you share your generations!
@nickfloats
The rate of change is reaching incredible speed, The overwhelming majority of the population don’t even know what Midjourney is! Over that same year we have seen the birth of AI-generation in video, special FX, sound FX, Music and now video games as well as LLM’s . Future shock…
WE CAN GENERATE OUR OWN POP STARS WITH AI! The Song, the video and the artist were all created with generative AI. Please allow me to present Uncanny Harriet’s first song “Bits of Code”. Please stick with it as it doesn’t hit its stride until about 0.45. Tools used…
Top frame is the input video, the middle is the picture I used as a prompt and the bottom is the
#gen1
output. That’s it, no prompt engineering just the picture.
@runwayml
is changing the course of video production and it’s just getting started. Music by AlexiAction from Pixabay
I made an Anime with AI! Here is my latest Ai film it's called "A Girl and Her Cat". It a post apocalyptic Anime, trust me, you're going to what the sound on for this one. Moonvalley, Pika and Midjourney for the visuals. Splash and Eleven Lab for the music and voices. Let me know…
I made officer Fang in midjourney and brought her to life with some stock footage and fair bit of kitbashing on this one:
@ViggleAI
@runwayml
@KaiberAI
@FaceFusionApp
and
@elevenlabsio
... remember what we kit bashers do now, is an off the shelf product in 6/8 months... who needs…
I am sharing this tech demo/workflow test I set myself, I can't resist a challenge! It's very rough round the edges as I didn't want to invest too much time (or Credits!) into something which was just an experiment. I improvised a scene with Chatgpt playing the role of Helen…
Ok I've refined my AI Talkie workflow, it's simpler, quicker and produces higher quality results! Here is a teaser below using the voice from my last test, X does not do it justice, you can see it glorious 4K on my youtube channel. After this weekend I will make my next full…
Runway has pulled off an amazing update and given us control of the movement and camera controls when we extend our generations! Watch below with lipsynch by Lalamu and added Sound… we can now pull off truly cinematic single take shots!!! Props to
@Kabooki_Kai
for spotting the…
Sound on please for my latest AI generated cinematic showreel. Made with Midjourney, Runway Gen-2, Magnific and Adobe CS. Proud to be a member of the
@runwayml
creative partnership,
@Magnific_AI
. 4k available on my YouTube soon, link in Bio.
Created these characters in Leonardo using the new transparency mode then animated them in Viggle, just need to be upscaled and placed in a scene now, will play with this more when I get time 👀
@LeonardoAi_
@ViggleAI
Just noticed I’ve hit 7000 followers on here. Thank you for giving me your attention. Think I might of discovered a way of using AI to get more followers on this site 😂 Sound on 🔉
This one’s for the Lord of the Rings fans. Sound on subtitles available and, as always, everything you see and hear is AI generated…
#LOTR
@midjourney
@pika_labs
Here is mine and
@ai_artworkgen
submission for Gen48. Images generated on Midjourney, sound and music epidemic
@epidemicsound
, voice
@elevenlabsio
edited on
@adobe
. It was great working with Marco and this was a real collaborative effort. The OCD in me wants to do another couple…
How has Gen-2’s marriage with
#midjourney
advanced the fundamentals of AI cinema? Things like consistent characters and locations? Changing shot length, sizes, angles & emotions? Everything you see and hear in this test is AI generated. Sound on and stay to the end 👀
@runwayml
Using all the tools to ensure consistency of character on my latest AI short. The original footage on the left is
@runwayml
text2video. On the right I’ve done a face replace and then used photoshop gen fill to generate a new shirt and tie and pasted it on with ebsynth. It’s not…
Little test for my AI animated short film. Images made in
#midjourney
with AI voice replacement, dialogue by Chatgpt, animated, edited graded on iPhone…. I think I can make it work!
#midjourneyart
#AIArtwork
Latest little test is a mix of text/image prompts and pre-set styles that chang only me and not the background. All 10 second generations to test for consistency. This is fast becoming one of my favourite style transfer models
@lensgo_ai
Here is my showreel. It's built from video I generated with AI over the last 3 months. I used Gen1, Gen2, KaiberAI, Midjourney, VoiceAI and Bark. Please put the sound on, like, share and give us a follow for more insight into AI generated video. Cheers!
Hope no one over did it at the weekend 👀 Test of a pure AI workflow, No video used in this at all. Midjourney for the background and character design,
@Mootion_AI
for the animation,
@ViggleAI
for the character style transfer and
@KaiberAI
to bring it all together. Elevenlabs for…
Made this Viggle dialogue test in about 20 minutes on my mobile in the cab ride home from work. Ignore the abysmal dialogue and acting I shot it in 20 seconds waiting for the cab 😂.
@ViggleAI
can do lip synch too 👀 props to
@francolli
for the discovery! Opens up new AI…
The promised land of consistent characters in the same scene, with changing emotions can now be achieved with generative fill! (despite being bloody fiddly) That means it can be achieved in Gen-2 and Pika too. So we can get traditional blocking and coverage for our scenes. Great…
Here's a thread of some of the tracks I have been making with Udio the incredible new AI Music Generator: First one is probably Uncanny Harriet's next video.. Up until:
@udiomusic
Hit the over 5000 followers today, good night to you one and all. The king of nightmares awaits 👹…. Made with
@runwayml
’s new pan and tilt camera controls,
@Magnific_AI
@
@midjourney
and
@adobe
, sound on. Thanks for following 👍
How we all feeling this morning post-Sora? A little uncertain about the future? Overwhelmed by the speed of change? Now is the time to focus on your storytelling skills, writing, world-building, direction, editing, sound design. Sora won’t be the only tool to deliver this level…
One of the most incredible features of AI is the use of style transfers to reimagine existing work. You can remix any video, even AI generated video, to build new content from old. It really encapsulates the Interdimensional Cable, Everything everywhere all at once, Multiverse of…
So to my surprise I’ve passed 3000 followers. I made this little Gen-2 generation to say thank you for following me on my AI adventures! Song: Falling Down Music by: Here are some other AI creator accounts worthy of a lot more attention. Give ‘em a follow:…
Great to see one of my first AI generated videos featured on
@BBCClick
this week. Twitter deleted it when I tried to make it a highlight but you can watch it below. I even got to meet
@spenley
Finally starting to see Video Generation filtering into mainstream.
Here's my latest AI generated video, another mock advertisement this time for a fashion brand in the cyberpunk 2077 universe. See it in glorious 4k here:
While your there I'd really appreciate it if you would also subscribe to my Youtube channel as I will…
As a test with new AI video generation models I try to resurrect a certain individual using the model and AI deepfakes. Sound on and a wait for it. Everything you see and hear (a part from the voice, due to a faulty microphone) is AI.
Proof of concept: using Gen-2 with the extend feature and Wav2lip to bring my uncanny_Harry avatar to life. Rough test but it works. He will introduce and present my videos on YouTube. Sound on please 👇🏻
With Mootion and now magic-animate we are entering a new era of character movement control. magic animate test below 👇🏻Image+inputvideo+outputvideo with in phone grade… Harry’s on the run! Try it on replicate and thanks to
@fofrAI
for the heads up!
Introducing Sora, our text-to-video model.
Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions.
Prompt: “Beautiful, snowy…
Just a little post to celebrate hitting 6000 followers.. although I might lose a few after posting this! 👀 Night… sound on…. Don’t let the bedbugs bite! Made with
@midjourney
@runwayml
@LeonardoAi_
@pika_labs
I have talked of seeing a pathway to making a a full length feature film using image/text2video in the next 8 months or so, I now see a freeway, today. It’s inevitable The game has radically changed, we now have directorial control over Gen-2
I don’t think most people understand the tsunami of creativity and content heading our way because of
@runwayml
new image2video and
@pika_labs
hitting open release. Brace for impact everybody it’s going to be amazing!
My Film Cold Call has been shortlisted for the Budapest AI Film Festival. Great to see festivals enabling the collective experience of viewing AI films!
Had an interesting conversation with an established television producer looking to pivot into AI production with a new start up. The starting gun has been fired, at the end of this year into next year, you’ll be seeing a lot of mainstream Music Videos and Commercials featuring AI…
Runway has dropped a new lip-synch tool for Creative Partners to test out and it's bloody brilliant
@runwayml
Voice and Sound effects by the amazing
@elevenlabsio
Final test before I make my full AI short on
#Gen2
. Everything you see and hear is totally AI generated. In this test I am putting together the stuff I’ve learned about creating a consistent character in a consistent space that can show a range of emotions. While still testing…
Spent a hour or so playing around with some stock presentation footage to test the latest AI faceswap, Voice replacement and lipsynch tech. It's not perfect but it's defiantly at the "good enough to convince people on facebook" level. Sound on and watch as I peel back the onion.…
In the not too distant future…
Me: Yeah, I want to watch an action film set in the Blade Runner Universe, directed by John Woo and starring me.
My AI: Generating…
Remember peeps AI film is for everyone, for every genre, for every taste, it's not just robots, anime and astronauts as illustrated in this beautiful piece by Chrissie.
Gm friends! Last night I played with a beautiful prompt from
@PhletchyIRL
. I finally realized the images were reminding me of the beginning of a song I love so I created a video.
Enjoy and try out the prompt (comments)
Niji V6 is a beast! Just need zoom and vary region to make my Cyberpunk Anime short! Come on
@midjourney
Made with
@LeonardoAi_
motion and
@pika_labs
Good night all AI peeps! Music by Vivek
Here's a test using a long take with lip-synch in generative AI video. I used
@runwayml
to animate a
@midjourney
created image. I went for a fixed camera (speed 0) motion 2, all first rolls and extended, I could of improved performance with re-rolls but this is just a quick test.…
Top tip for
#Gen2
by
@runway
. At present it doesn’t cope with a lot happening in one shot, it will take quite a few generations to get a useable 2-shot. So you need to use the Kuleshov effect, to cut between shots to create meaning. Watch the scene below and see how the tension…
WIP update: Reaching the end of my project today and I’ve learned so much that I have unlocked an even better work flow for the next project. Feel like the age of AI film making has truly arrived and next year it will explode into the mainstream. We’ve got consistent characters,…
Check out Paul’s amazing Sora video 🤯. Paul was one of the first film makers to access
@runwayml
and now one of the first to play with Sora. It’s the second time he has made a small clique of nerds envious 😂 Great to see creatives making stuff with Sora, now things get…
Made with Sora. The Golden Record - from raw earth material to a time capsule of human life on Earth. Using 11 different generations cut together from Sora, I was able to explore what the odyssey of this record might look like.
@OpenAI
I’ve been playing with a brilliant new AI text-to-video generator
@pika_labs
and I made this cursed short film about the power of nightmares with it. Sound/Headphones on please! If you want to know the tools I used to make it, stick around for the end credits. Thanks to the…
The motion brush has landed at a perfect time for my project. Take a look at this 17 second shot with a head turn and a subtle zoom in.
@runwayml
making us all directors now with this level of control…
Testing out the new
@lensgo_ai
style transfer feature with some free credits they gave me. Load up a video and change it with AI. You can can let the AI effect everything or just people as I have done here. I am not wearing a Spiderman suit in my living room, it's AI! This is one…
New bench mark in lipsynching, amazing! Congratulations to the team. 👏👏👏 My test, image in
@midjourney
, animated with
@runwayml
, voice made with
@elevenlabsio
with a little upscale below 👇🏻
Video ReTalking Colab 🥳 Thanks to Kun Cheng ❤
@shadocun
❤ Yong Zhang ❤ Menghan Xia ❤ Fei Yin ❤ Mingrui Zhu ❤ Xuan Wang ❤ Jue Wang ❤ Nannan Wang ❤
📄paper:
🧬code:
🦒colab: please try it 🐣
Workflow Pop Quiz: I was challenged with the task of demonstrating new AI video tools in a single take piece to camera by the BBC click team. Although it wasn’t used as a single take on the Final Cut for editoral reasons I was pleased with the result. Can you name the tools used?
Little lip synch test with
#gen2
using rough AI voices. Entirely AI actor speaking lines of dialogue. It’s rough around the edges but I already know how to vastly improve it.
#synthography
Pika works really well with fantasy and animation generations. Anyone up for a dark Pixar universe? Kicking off with Taxi Driver? So wrong but so right 😂 Sound on please
@pika_labs
👇🏻
Feeling pleased with myself as I’ve just handed in a fun project which I won’t be able to talk to about for awhile. Now I chill with the family just in time for the weekend. Have a great one AI peeps 🥃 Made with
@runwayml
Gen2 image2video and lip-synch
@elevenlabsio
for sound…
AI film making doesn’t have to try to mimic reality, it excels in being weird, avant-garde, dreamlike or even nightmarish. In that spirit here is my latest short “Sick Puppy” and if you like stuff from left field may I recommend
@rainisto
@paultrillo
@loved_orleer
@katiemorrisart
…
Happy New Year to all my friends and followers on X The AI community on here has been an amazing source of information and support this year. 2024 is going to be a mad one! Fireworks supplied by
@runwayml
👇
Thrilled to have been accepted onto the creative partners program with
@runwayml
. It’s a privilege to have a front row seat witnessing the incredible innovation taking place. Excited to continue pushing the boundaries of generative video AI. The journey’s just beginning! 🎥🚀
Played with
#Bark
today it hasn’t got the clarity of elevenlabs but more than makes up for it in emotional range and authenticity and it’s free! Thanks to
@fofrAI
for the tutorial , check out the voice I generated with a
#gen2
avatar below. Sound on 👇🏻
#synthography
Sharing the scene I made for the
@t2remake
way back in October, scene 29. Made with
@runwayml
Gen1 with midjourney generated prompts, faceswaping and pre
@elevenlabsio
Voice2voice AI. So much fun to make and a great community experience where I met some fantastic people. Big up…
I mini thread and first impressions of
@KaiberAI
new text2video mode. Kaiber continues to innovative with a slider controller that influences the amount of movement in your generation and previews that give you an indication of the type of movement to expect, all delivered with…
I just want to remind everyone that I am not paid to promote any products. All the tools I post about I use in my own work. So when I say that
@elevenlabsio
is an incredible tool, know that I mean it. Big up the tool makers!
@fofrAI
@venturetwins
Every week my mum sends me something to ask if its' AI. 100% of the time it's AI. Last week she sent me a advert for a product that could help my dodgy shoulder complete with a badly Wav2liped doctor explaining it's virtues. ME: Look at the lips mum, that's AI too! Boomers on…