TomLikesRobots🤖's banner

TomLikesRobots🤖

@TomLikesRobots • 41,824 subscribers

Filmmaker at Fable Studios. AI Specialist. IMDB link below. Previously AI VFX at Metaphysic. Featured in Forbes, Hollywood Reporter, The Guardian, New Yorker.

Shorts

Experimenting with a watercolour animation style on SeeDance 2 and Adrien Brody appears on the TV (with canned laughter?). Does anyone else get actors/celebrities randomly turning up in their videos?

Experimenting with a watercolour animation style on SeeDance 2 and Adrien Brody appears on the TV (with canned laughter?). Does anyone else get actors/celebrities randomly turning up in their videos?

21,328 次观看

Apparently SeeDance 2 is down for the time being on BytePlus but here's a young wizard with his dragon companion. It's a very cool video model. This video cost $0.72 USD to generate which seems reasonable. (300,000 tokens for a 15 secs at $0.0024 USD per 1000 tokens.)

Apparently SeeDance 2 is down for the time being on BytePlus but here's a young wizard with his dragon companion. It's a very cool video model. This video cost $0.72 USD to generate which seems reasonable. (300,000 tokens for a 15 secs at $0.0024 USD per 1000 tokens.)

50,451 次观看

Cool. There's a new AI video lip sync tool available. Omnihuman is now accessible as a product via Capcut's Dreamina website and you can generate a 15 second trial video for free when you sign up. Mouth articulation and general movement look really good. I need to play with this a bit more.

Cool. There's a new AI video lip sync tool available. Omnihuman is now accessible as a product via Capcut's Dreamina website and you can generate a 15 second trial video for free when you sign up. Mouth articulation and general movement look really good. I need to play with this a bit more.

142,564 次观看

Gen-4 is so much fun. I have a series of Midjourney generations in this miniature diorama style that no other model has been able to animate. They freeze all the figures or if they do move totally change the style. Gen-4 interprets beautifully in terms of movement and maintaining style. At last, I may be able to create this short I've been thinking about for a while.

Gen-4 is so much fun. I have a series of Midjourney generations in this miniature diorama style that no other model has been able to animate. They freeze all the figures or if they do move totally change the style. Gen-4 interprets beautifully in terms of movement and maintaining style. At last, I may be able to create this short I've been thinking about for a while.

83,383 次观看

I'm enjoying playing with Sora 2. It feels more like a meme generator than a serious tool so far, but it is great fun. If anyone needs an invite code, let me know.

I'm enjoying playing with Sora 2. It feels more like a meme generator than a serious tool so far, but it is great fun. If anyone needs an invite code, let me know.

42,616 次观看

I've just received access to SeeDance2. Very first test and it really does live up to its hype. This is a slow and considered generation but it is very, very cool. Down with a virus right now, but I'll enjoy testing this.

I've just received access to SeeDance2. Very first test and it really does live up to its hype. This is a slow and considered generation but it is very, very cool. Down with a virus right now, but I'll enjoy testing this.

18,275 次观看

This could be huge for AI storytelling. I've tried so many AI lipsync tools and have never been satisfied. This is my first try with Hedra Character-3 and it's blown me away. It was so easy. I uploaded a track from Suno and an image from Midjourney and it just worked.

This could be huge for AI storytelling. I've tried so many AI lipsync tools and have never been satisfied. This is my first try with Hedra Character-3 and it's blown me away. It was so easy. I uploaded a track from Suno and an image from Midjourney and it just worked.

42,696 次观看

“I ate his liver with some fava beans and a nice chianti.” It took a bit of time to match up the mouth and eye movements and needed a few keyframes but I think the videos blend together pretty well. 0.6 Denoise. #aiart #stablediffusion #EbSynth #controlnet

“I ate his liver with some fava beans and a nice chianti.” It took a bit of time to match up the mouth and eye movements and needed a few keyframes but I think the videos blend together pretty well. 0.6 Denoise. #aiart #stablediffusion #EbSynth #controlnet

38,733 次观看

Quick tip: You probably know that you can extend an AI video by extracting the last frame and using that frame as the initial image for image to video. If you join the videos with Capcut (free video editor) you can use Mix to smooth out the transition which will very quickly improve your result. AI videos here generated with Pika (super impressive text to video) and Minimax (super impressive image to video).

Quick tip: You probably know that you can extend an AI video by extracting the last frame and using that frame as the initial image for image to video. If you join the videos with Capcut (free video editor) you can use Mix to smooth out the transition which will very quickly improve your result. AI videos here generated with Pika (super impressive text to video) and Minimax (super impressive image to video).

18,816 次观看

While we're waiting for the full Midjourney V7 to drop, here's a V7 image animated with Runway Gen-4 with sound from MMAudio.

While we're waiting for the full Midjourney V7 to drop, here's a V7 image animated with Runway Gen-4 with sound from MMAudio.

12,437 次观看

It's good to see that #pix2pix has been integrated into #Automatic1111. img2img detects when pix2pix checkpoint is selected and so extension no longer needed. Now possible to use XYZ plots to find perfect settings. Make it Winter: from 0-30 txt cfg #stablediffusion #aiart

It's good to see that #pix2pix has been integrated into #Automatic1111. img2img detects when pix2pix checkpoint is selected and so extension no longer needed. Now possible to use XYZ plots to find perfect settings. Make it Winter: from 0-30 txt cfg #stablediffusion #aiart

17,803 次观看

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

SeeDance2 is now available on Runway. It's an amazing model and Runway's interface is the perfect place to experiment. Interestingly the soundtrack was generated by SeeDance and I extended it using Suno. SoundFX were also added and they provided continuity across the video.

SeeDance2 is now available on Runway. It's an amazing model and Runway's interface is the perfect place to experiment. Interestingly the soundtrack was generated by SeeDance and I extended it using Suno. SoundFX were also added and they provided continuity across the video.

TomLikesRobots🤖

133,678 次观看 • 1 个月前

This is the start of an AI animated mood film I've been working on over the weekend with input from my nine year old. She loves anime and sat with me saying what she liked and what she didn't. It's a collection of shots mostly using Midjourney Niji to Luma Ray 2, image to video. Audio from MMAudio and Suno.

Sensitive content

This media may contain sensitive content.

This is the start of an AI animated mood film I've been working on over the weekend with input from my nine year old. She loves anime and sat with me saying what she liked and what she didn't. It's a collection of shots mostly using Midjourney Niji to Luma Ray 2, image to video. Audio from MMAudio and Suno.

TomLikesRobots🤖

321,691 次观看 • 1 年前

SeeDance 2.0 vs HappyHorse 1.0 Very quick text-to_video comparison. Which do you prefer? Universal Prompt: "Aesthetic: lofi, cozy, warm tones, soft film grain, gentle bloom, analog textures. Camera default: 35 to 50mm equivalent, shallow depth of field, slow subtle moves, cut length 0.6 to 1.2 seconds. Style: Cozy cel-shaded anime." Thanks to PixVerse for HappyHorse generations which are currently free to members. lofi audio tracks were pretty poor for both so I used Suno for audio.

SeeDance 2.0 vs HappyHorse 1.0 Very quick text-to_video comparison. Which do you prefer? Universal Prompt: "Aesthetic: lofi, cozy, warm tones, soft film grain, gentle bloom, analog textures. Camera default: 35 to 50mm equivalent, shallow depth of field, slow subtle moves, cut length 0.6 to 1.2 seconds. Style: Cozy cel-shaded anime." Thanks to PixVerse for HappyHorse generations which are currently free to members. lofi audio tracks were pretty poor for both so I used Suno for audio.

TomLikesRobots🤖

14,648 次观看 • 1 个月前

"The Anomaly of Kepler-61b" Entirely generated with AI. Script: #ChatGPT4 txt2video: #Gen2 txt2speech: I'm loving being able to spend a couple of hour building up a story with the help of AI. It feels like a taste of the future and these tools will only get better. #aiart #GenerativeAI

"The Anomaly of Kepler-61b" Entirely generated with AI. Script: #ChatGPT4 txt2video: #Gen2 txt2speech: I'm loving being able to spend a couple of hour building up a story with the help of AI. It feels like a taste of the future and these tools will only get better. #aiart #GenerativeAI

TomLikesRobots🤖

205,613 次观看 • 3 年前

So much potential with this combination. I really think it would be possible to make a cosy animation people would watch. Midjourney V7 Runway Gen-4 Suno AI

So much potential with this combination. I really think it would be possible to make a cosy animation people would watch. Midjourney V7 Runway Gen-4 Suno AI

TomLikesRobots🤖

90,552 次观看 • 1 年前

Until recently I disliked using upscalers on video and their mushy or oversharpened looks. Topaz Starlight 2.5 and Magnific Video are now available and both give great results and improve the look of 720p SeeDance videos, even when upscaling to 1080p. Topaz seems to over-enhance skin blemishes in this example and Magnific seems slightly more natural but all blemishes have gone. There are plenty of parameters to play with in Magnific so perhaps with playing I could get better results. Does anyone have a preference? Interested in seeing some more comparisons? Magnific does especially well with flat 2D animation. Maybe there's more of a difference at higher resolutions.

Until recently I disliked using upscalers on video and their mushy or oversharpened looks. Topaz Starlight 2.5 and Magnific Video are now available and both give great results and improve the look of 720p SeeDance videos, even when upscaling to 1080p. Topaz seems to over-enhance skin blemishes in this example and Magnific seems slightly more natural but all blemishes have gone. There are plenty of parameters to play with in Magnific so perhaps with playing I could get better results. Does anyone have a preference? Interested in seeing some more comparisons? Magnific does especially well with flat 2D animation. Maybe there's more of a difference at higher resolutions.

TomLikesRobots🤖

15,730 次观看 • 1 个月前

Marble from World Labs is so close to magic. You can imagine a space, generate it and then walk around it in your VR Headset. This was recorded with my Quest 3 and the potential is mind-blowing.

Marble from World Labs is so close to magic. You can imagine a space, generate it and then walk around it in your VR Headset. This was recorded with my Quest 3 and the potential is mind-blowing.

TomLikesRobots🤖

17,460 次观看 • 8 个月前

没有更多内容可加载