CoffeeVectors's banner
CoffeeVectors's profile picture

CoffeeVectors

@CoffeeVectors • 42,297 subscribers

AI Filmmaker // AI Creative Director at https://t.co/AXEILADYA8

Shorts

First test with the US release of Seedance 2. Wanted to see how it handles messier illustrated fight scenes.

First test with the US release of Seedance 2. Wanted to see how it handles messier illustrated fight scenes.

63,545 views

Asked o3-mini to make a quick water sim with an orbiting camera in Blender 4.3 using a python script. This took a few tries and some tweaking so it’s not a one shot. I’m optimistic this will improve down the road. Prompt below.

Asked o3-mini to make a quick water sim with an orbiting camera in Blender 4.3 using a python script. This took a few tries and some tweaking so it’s not a one shot. I’m optimistic this will improve down the road. Prompt below.

314,054 views

Another Seedance 2 test! Handheld running shot with an overflowing bag of gold coins, jumping into water, and having the camera follow him underwater. Can’t wait until this model is fully released! Prompt below 👇

Another Seedance 2 test! Handheld running shot with an overflowing bag of gold coins, jumping into water, and having the camera follow him underwater. Can’t wait until this model is fully released! Prompt below 👇

77,171 views

Testing machine animations in Midjourney video.

Testing machine animations in Midjourney video.

92,408 views

boom. // Seeing how Seedream 4 on @FAL handles ultrawide closeups on a consistent character. Very impressed by the extreme angles it can generate! Keyframing with Hailuo AI (MiniMax) 🧵Music in @SunoMusic Base image Midjourney

boom. // Seeing how Seedream 4 on @FAL handles ultrawide closeups on a consistent character. Very impressed by the extreme angles it can generate! Keyframing with Hailuo AI (MiniMax) 🧵Music in @SunoMusic Base image Midjourney

73,365 views

Used Pika 2’s new ingredients feature to make the top video. Then did vid2vid with it in Sora (1080p Mild Remix). Second clip is the same videos, just zoomed in. Ingredients in 🧵

Used Pika 2’s new ingredients feature to make the top video. Then did vid2vid with it in Sora (1080p Mild Remix). Second clip is the same videos, just zoomed in. Ingredients in 🧵

77,239 views

Running #controlnet tests on scenes with dialog. Here I was seeing how an anime-watercolor vibe at 12 fps would track on this bit from Anchorman. Had a lot of issues with rendering visemes. More research! #stablediffusion #artgen #aicinema #aiia

Running #controlnet tests on scenes with dialog. Here I was seeing how an anime-watercolor vibe at 12 fps would track on this bit from Anchorman. Had a lot of issues with rendering visemes. More research! #stablediffusion #artgen #aicinema #aiia

70,178 views

Test using an iPad as #virtualcamera in #UnrealEngine5 with a #metahuman rendered in #stablediffusion Multi-ControlNet. Deflickering in #DaVinciResolve. Much to improve & research, but wanted to share! #aicinema #controlnet #aiia Unreal Engine #aiphotography

Test using an iPad as #virtualcamera in #UnrealEngine5 with a #metahuman rendered in #stablediffusion Multi-ControlNet. Deflickering in #DaVinciResolve. Much to improve & research, but wanted to share! #aicinema #controlnet #aiia Unreal Engine #aiphotography

60,164 views

Testing fight scenes using Midjourney -> Hailuo AI (MiniMax) Motion blur added and reduced framerate to 16fps in post. Video Prompt: anime fight scene, extremely dynamic camera, acrobatics, punches, kicks, fast pace action

Testing fight scenes using Midjourney -> Hailuo AI (MiniMax) Motion blur added and reduced framerate to 16fps in post. Video Prompt: anime fight scene, extremely dynamic camera, acrobatics, punches, kicks, fast pace action

21,097 views

Videos

LIVE
1.2k

Anya Rossi

sweetdream.ai

SweetDream.aiSponsored•Livecam
Streaming Now

Watch Anya Live

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

HD live stream
Exclusive private shows
1.2k viewers online

Free preview available • Premium content

A Gift From Your Mother // Seedance 2
0:22
CoffeeVectors's profile picture

A Gift From Your Mother // Seedance 2

CoffeeVectors

651,037 views • 1 month ago

Descent // Seedance 2 Runway
0:34
CoffeeVectors's profile picture

Descent // Seedance 2 Runway

CoffeeVectors

76,787 views • 28 days ago

Timelapse of using #photoshop’s new generative fill feature to connect two images and build a scene around them using blank prompts. Was inspired by MatthieuGB’s post doing something similar! Notice how I’m not adding any descriptions, but letting gen fill present options for extending the current selection as I play with different aspect ratios to extend the frame in steps. Added some music. Images follow🧵Adobe Adobe Photoshop 1/4
0:48
CoffeeVectors's profile picture

Timelapse of using #photoshop’s new generative fill feature to connect two images and build a scene around them using blank prompts. Was inspired by MatthieuGB’s post doing something similar! Notice how I’m not adding any descriptions, but letting gen fill present options for extending the current selection as I play with different aspect ratios to extend the frame in steps. Added some music. Images follow🧵Adobe Adobe Photoshop 1/4

CoffeeVectors

1,725,674 views • 3 years ago

Made this video with iPhone photos I took of my friend Stephanie that I used as keyframes in Luma! With the camera controls I can gen transitions between shots. I also built a custom web app in Next.js to help me speedramp and edit all the clips! Breakdown 🧵(1/18)
0:38
CoffeeVectors's profile picture

Made this video with iPhone photos I took of my friend Stephanie that I used as keyframes in Luma! With the camera controls I can gen transitions between shots. I also built a custom web app in Next.js to help me speedramp and edit all the clips! Breakdown 🧵(1/18)

CoffeeVectors

338,191 views • 1 year ago

Tests using a #metahuman from #UnrealEngine5 with #stablediffusion Multi-ControlNet. Applied diff. amounts of deflickering in #DaVinciResolve. Mainly seeing how it handles 3D camera moves. #aicinema #controlnet #aiia Unreal Engine #aiphotography #MachineLearning #DeepLearning
0:58
CoffeeVectors's profile picture

Tests using a #metahuman from #UnrealEngine5 with #stablediffusion Multi-ControlNet. Applied diff. amounts of deflickering in #DaVinciResolve. Mainly seeing how it handles 3D camera moves. #aicinema #controlnet #aiia Unreal Engine #aiphotography #MachineLearning #DeepLearning

CoffeeVectors

446,088 views • 3 years ago

Testing LCM LORAs in an AnimateDiff & multi-controlnet workflow in ComfyUI. I was able to process this entire Black Pink music video as a single .mp4 input. The LCM lets me render at 6 steps (vs 20+) on my 4090 and uses up only 10.5 GB of VRAM. Here's a breakdown 🧵[1/11]
3:25
CoffeeVectors's profile picture

Testing LCM LORAs in an AnimateDiff & multi-controlnet workflow in ComfyUI. I was able to process this entire Black Pink music video as a single .mp4 input. The LCM lets me render at 6 steps (vs 20+) on my 4090 and uses up only 10.5 GB of VRAM. Here's a breakdown 🧵[1/11]

CoffeeVectors

182,286 views • 2 years ago

Finally got #gaussiansplatting working! I’m really impressed by how well it captured the fine details of the sneaker fabric and shoelaces, as well as the text. This was generated from 75 DSLR photos I shot on a 24mm lens, took about 30 minutes to process on an RTX 3090, and is running at a smooth 60fps on the SIBR Viewer. Going to run more tests to see how this compares to some of my experiments in photogrammetry. I’m hoping a plugin for #UnrealEngine5 comes out soon! #nerf
0:21
CoffeeVectors's profile picture

Finally got #gaussiansplatting working! I’m really impressed by how well it captured the fine details of the sneaker fabric and shoelaces, as well as the text. This was generated from 75 DSLR photos I shot on a 24mm lens, took about 30 minutes to process on an RTX 3090, and is running at a smooth 60fps on the SIBR Viewer. Going to run more tests to see how this compares to some of my experiments in photogrammetry. I’m hoping a plugin for #UnrealEngine5 comes out soon! #nerf

CoffeeVectors

183,376 views • 2 years ago

Destiny // Testing FPV prompts on WWI battlefields in Kling AI Prompt example 👇
0:47
CoffeeVectors's profile picture

Destiny // Testing FPV prompts on WWI battlefields in Kling AI Prompt example 👇

CoffeeVectors

33,954 views • 4 months ago

Really impressed by the acting you can generate with #texttospeech! These are #AIvoices generated from text—dozens of "takes" stitched together. Breakdown thread: 1/8 #syntheticvoices #ainarration #autonarrated #aicinema #aiartcommunity #aiia #ai #MachineLearning
0:51
CoffeeVectors's profile picture

Really impressed by the acting you can generate with #texttospeech! These are #AIvoices generated from text—dozens of "takes" stitched together. Breakdown thread: 1/8 #syntheticvoices #ainarration #autonarrated #aicinema #aiartcommunity #aiia #ai #MachineLearning

CoffeeVectors

192,104 views • 3 years ago

More Wan 2.2 tests on Here I wanted to compare it directly to Veo 3 and also see how tight into the eye I could get with complex zooms. Again I’m really surprised by what Wan 2.2 can do. The prompts are the same across models, so Wan seems to be able to do more with the text, at least with camera moves and optics. 🧵
0:53
CoffeeVectors's profile picture

More Wan 2.2 tests on Here I wanted to compare it directly to Veo 3 and also see how tight into the eye I could get with complex zooms. Again I’m really surprised by what Wan 2.2 can do. The prompts are the same across models, so Wan seems to be able to do more with the text, at least with camera moves and optics. 🧵

CoffeeVectors

60,690 views • 9 months ago

Made this video (🎶) with a Midjourney v6 image! Started by upscaling/refining with Magnific.ai, pulled a Marigold Depth Map from that in ComfyUI, then used as a displacement map in Blender where I animated this camera pass with some relighting and narrow depth of field.🧵1/12
0:55
CoffeeVectors's profile picture

Made this video (🎶) with a Midjourney v6 image! Started by upscaling/refining with Magnific.ai, pulled a Marigold Depth Map from that in ComfyUI, then used as a displacement map in Blender where I animated this camera pass with some relighting and narrow depth of field.🧵1/12

CoffeeVectors

135,627 views • 2 years ago

Skool Daze A little punk & mixed media in my latest Midjourney video! Interesting how dynamic the camera can get with some of the shots. @SunoMusic
1:45
CoffeeVectors's profile picture

Skool Daze A little punk & mixed media in my latest Midjourney video! Interesting how dynamic the camera can get with some of the shots. @SunoMusic

CoffeeVectors

62,343 views • 11 months ago

More tests with #controlnet using a scene from #terminator2. Was going for a watercolor/ink vibe. Top is 24 FPS. Bottom is 8 FPS with more of a graphic novel/animatic theme. #AImusic generated by Breakdown thread—1/10 #aicinema #stablediffusion #aiia
0:57
CoffeeVectors's profile picture

More tests with #controlnet using a scene from #terminator2. Was going for a watercolor/ink vibe. Top is 24 FPS. Bottom is 8 FPS with more of a graphic novel/animatic theme. #AImusic generated by Breakdown thread—1/10 #aicinema #stablediffusion #aiia

CoffeeVectors

115,865 views • 3 years ago

brain rotting, but the thoughts still racing // Testing img2vid lipsyncing with the InfiniteTalk model on WaveSpeedAI and a song I made in Suno v5. Love how it handles images with multiple faces and how it animates even when it doesn’t see a mouth. Quick tutorial in 🧵👇 (1/6)
0:53
CoffeeVectors's profile picture

brain rotting, but the thoughts still racing // Testing img2vid lipsyncing with the InfiniteTalk model on WaveSpeedAI and a song I made in Suno v5. Love how it handles images with multiple faces and how it animates even when it doesn’t see a mouth. Quick tutorial in 🧵👇 (1/6)

CoffeeVectors

29,881 views • 6 months ago

Testing a continuous walking shot through different areas in Veo 3. Music in @SunoMusic
0:28
CoffeeVectors's profile picture

Testing a continuous walking shot through different areas in Veo 3. Music in @SunoMusic

CoffeeVectors

36,054 views • 10 months ago

Trying Ethan Mollick’s Veo 3 prompt but with movies as community theater. These are pretty fun! Also surprised AI knows how to do this. Avengers, Lord of the Rings, Transformers, Harry Potter, Top Gun, Dune, Predator, Ghostbusters, Titanic, Wakanda Forever, Batman, 300
1:16
CoffeeVectors's profile picture

Trying Ethan Mollick’s Veo 3 prompt but with movies as community theater. These are pretty fun! Also surprised AI knows how to do this. Avengers, Lord of the Rings, Transformers, Harry Potter, Top Gun, Dune, Predator, Ghostbusters, Titanic, Wakanda Forever, Batman, 300

CoffeeVectors

34,477 views • 10 months ago

Nothing at All // Testing skateboard shots with Midjourney video. Music @SunoMusic
1:16
CoffeeVectors's profile picture

Nothing at All // Testing skateboard shots with Midjourney video. Music @SunoMusic

CoffeeVectors

32,917 views • 10 months ago

Testing Wan 2.2 on It’s able to do cool stuff with camera that I can’t seem to do with the other models. Its the dutch angles, the creamy rack focus, and it goes in for really tight wide angle closeups with extremely high detail. And it’s open source!
0:40
CoffeeVectors's profile picture

Testing Wan 2.2 on It’s able to do cool stuff with camera that I can’t seem to do with the other models. Its the dutch angles, the creamy rack focus, and it goes in for really tight wide angle closeups with extremely high detail. And it’s open source!

CoffeeVectors

27,964 views • 9 months ago

A range of motion test applying LCM LORA-HotshotXL-ControlNet to a Metahuman input. Here I used Zoe Depth and Realistic Line Control Nets, but was also experimenting with OpenPose, etc. Still trying to get more expression in the mouth for visemes & stabilize the hairline.
0:45
CoffeeVectors's profile picture

A range of motion test applying LCM LORA-HotshotXL-ControlNet to a Metahuman input. Here I used Zoe Depth and Realistic Line Control Nets, but was also experimenting with OpenPose, etc. Still trying to get more expression in the mouth for visemes & stabilize the hairline.

CoffeeVectors

42,959 views • 2 years ago

Testing Hedra new lip syncing Character 3 model on a song I made in Suno Ref images made in Magnific.ai and also upscaled for the closeups variations. Pretty incredible for for just img&audio2vid! Camera zooms/editing/color -grade/grain in post. Upscaled to 4k.
0:44
CoffeeVectors's profile picture

Testing Hedra new lip syncing Character 3 model on a song I made in Suno Ref images made in Magnific.ai and also upscaled for the closeups variations. Pretty incredible for for just img&audio2vid! Camera zooms/editing/color -grade/grain in post. Upscaled to 4k.

CoffeeVectors

22,944 views • 1 year ago