A.I.Warper @AIWarper Twitter profile

Pinned Tweet

A.I.Warper

5 days

Side by Side of my last post Obviously left is the driving video and right is my output Trying to find new things to test some of these newer performance transfer tools

4

5

90

Last Seen Profiles

@saraymnn

@Crypto_Briefing

@Bratzgem1

@TheBoysOOCC

@CarlyHughes101

@TheRealKano

@maisumamulher

@harudisney_0426

@_dongi_ii

@nunyabeeswaxfed

@mbdoy362Samara

@yuuki_cos_t

@Mirai_2402

@nikocoliterus

@FeelsXaadMan

@markhunt1974

@heartfuIz

@Sebastian04_21

@ButraguenoRM

@AntoniaE98

@biggersn15

@__rossie

@koujirou185

@Noel_Stryfe

@bubblegumpq

@jiratchaponng

@9dai_5

@iamAprilynv

@locopaphoto

@NLA_FB

@ATIQ_ALTHAGAFI

@iaragb

@aboomarMousa1

@purse_susann

@Kellfebruary7

@jiminparfk

A.I.Warper

@AIWarper

8 months

Here's probably one of the hardest shots I've ever attempted 1) Two characters 2) Two different race of humans (prompt bleeding) 3) Very fast paced with a lot of motion Was very fun and I learned a lot. Also develop a bunch of new workflow #StableDiffusion #Matrix #Pixar

1K

521

5K

A.I.Warper

@AIWarper

4 months

Viggle AI Experiment 💙 This is an incredible tool for style transfer. Details and link to Discord below 👇 As always, if you enjoy this content please kindly like and share!

185

803

4K

A.I.Warper

@AIWarper

8 months

💙Kendomland - land of the free and the men 🩷 This one was fun and feels unique I left the expressions and mouth movements more muted intentionally as it depicts a "doll" more clearly... I think? As always - please like and share if you enjoy this content! Love you all

102

433

3K

A.I.Warper

@AIWarper

6 days

Another example here of SAM V2 with literally 1 mouse on the first frame of the video and it tracked Simone even through that super fast motion part at the end If I were a VFX shop I'd be implementing this asap. It's better than any traditional ROTO I've ever used

50

220

2K

A.I.Warper

@AIWarper

8 months

Image to depth using Marigold and ConfyUI Then depth to 3D geometry in Blender Super interesting stuff 🤯

31

221

2K

A.I.Warper

@AIWarper

7 days

New SAM2 is an absolute game changer.... Did this track in 1 mouse click and about 4 secs. The fact that it even tracked across the scene change is mind blowing to me Will post some more tests below

27

169

2K

A.I.Warper

@AIWarper

2 months

Diffusion even set the mood to be more "jokeresque" !

28

173

1K

A.I.Warper

@AIWarper

9 months

Experimental workflow test 1) Rough Blender Animation Blockout 2) #RunwayML Gen 1 Text to Prompt for Base Style 3) #AnimateDiff for cleanup Rendered at 8 fps for time saving (just a test at this point) Interesting results #AIVideo #AIAnimation #StableDiffusion

27

119

963

A.I.Warper

@AIWarper

4 months

Somehow #adobe hasn't included gen fill into their After Effects product yet for videos... which BAFFLES me So I did it myself. Fully open source✅ All free tools✅ GG✅ Your move Adobe

51

83

940

A.I.Warper

@AIWarper

2 months

Object removal Wonder Dynamics vs my Comfy workflow Thoughts?

34

87

775

A.I.Warper

@AIWarper

8 months

Update on my experiments with a consistent full style transfer - the fast movements with blur are killing me. "toony / Pixaresque" look to Keanu. Things of note: 1) Belt changes color (I'd fix that in AE in post) 2) Some minor eye jitter at the end (fix in post) Fun stuff!

36

69

616

A.I.Warper

@AIWarper

7 days

Just wow..... absolutely amazing Balloon scene from up. Video desaturated so you can see the balloon I am tracking with SAM v2

21

57

612

A.I.Warper

@AIWarper

3 months

What if you could use AI as a Render Engine? 🤯 Check out this video where I replace Captain America with someone you may know 😉 (🔊 on) If you enjoy this content please consider liking and sharing - it really helps me grow! As always, details below 👇

18

63

582

A.I.Warper

@AIWarper

6 months

"No more forks?" Some thoughts in the comments 👇 Please kindly like an reshare if you enjoy this content it really helps me in the long run🩵

40

74

548

A.I.Warper

@AIWarper

8 months

Fixed the proportions on my Blender rig to not look like a skinny alien👽 Much better results! Only thing left to try would be to fine tune the animation to look more 2D hand drawn (animated on 2s) 🩵If you like these updates please like and share! It helps me grow 🩵

17

54

496

A.I.Warper

@AIWarper

5 days

Fully Automated object removal with SAM V2 Still at the mercy of the existing object removal models but when someone trains a good version of that it's lights out Mask didn't include smoke / car shadow so obviously it remains Not bad for a few mouse clicks

14

54

494

A.I.Warper

@AIWarper

4 months

Kijai the🐐 threw together a ComfyUI implementation 😊

9

88

459

A.I.Warper

@AIWarper

8 months

lil taste of what I've been working on Massive improvements to my style transfer with expression and lip sync carry over Please be sure to like and share as it helps me grow! #Animatediff #animation3d #AI #Stablediffusion

29

43

438

A.I.Warper

@AIWarper

5 months

"You'll have to talk louder, son!" Experiments with lipsync + expression transfer to a different medium. Some thoughts (Continued below as well) 1) I'm at the mercy of the available checkpoints. The "Disney / Pixar" models for 1.5 are very meh

24

46

400

A.I.Warper

@AIWarper

5 days

Remove the soccer ball and you are left with an oddly funny video 😂 Fully automated object removal with SAM V2 + ProPainter

6

47

434

A.I.Warper

@AIWarper

6 months

Optimistic about this one 👀

8

38

386

A.I.Warper

@AIWarper

4 months

I’ve succeeded to make a meme template so viral, the meme lord himself has made one!! 🥹🥹🥹🥹🥹🥹🥹🥹🥹🥹

DogeDesigner

@cb_doge

4 months

Elon Musk, when he posts a banger on 𝕏:

606

1K

10K

25

12

385

A.I.Warper

@AIWarper

6 months

Project textures in UnrealEngine with ComfyUI 🤯

10

70

367

A.I.Warper

@AIWarper

3 months

Super rough test of an animation using the new SDXL lineart ControlNET in my pipeline. Some obvious errors but it did well at capturing eye and mouth movements for such a larger medium shift! Promising stuff!

20

33

359

A.I.Warper

@AIWarper

7 days

Oddly worked much better than I expected it to lol SAM v2 tests on random videos

18

36

355

A.I.Warper

@AIWarper

1 month

Happy Independence day! 🇺🇸🇺🇸🇺🇸

19

43

349

A.I.Warper

@AIWarper

8 months

For people asking: No this is not achievable in one pass. You can’t just input the matrix frames, press go, this pops out the other side (yet). This was 8 different passes through workflows that took well over 100+ hrs to fine tune. Also a lot of roto work in AE for masks

64

9

341

A.I.Warper

@AIWarper

5 months

Kijai has implemented another method for extracting Depth / Normals through "Intrinsic Lora" The Normal map extraction is of particular interest as it is significantly more impressive that the existing preproc Link 👇

8

40

338

A.I.Warper

@AIWarper

1 month

AI will unlock all doors for creative people to transfer whats in their mind onto a canvas for all of us to enjoy without requiring: Millions in funding, artistic direction squashing from your boss, large studio bureaucracy, etc…

24

63

341

A.I.Warper

@AIWarper

6 months

Video 2 video god tier

12

31

330

A.I.Warper

@AIWarper

2 months

24 fps source 12 fps stylized

16

38

329

A.I.Warper

@AIWarper

4 days

I'm sure X will compress the heck out of this but..... Incredible detail coming out of this FLUX model. Truly incredible with only a second ksampelr pass for upscale and nothing else

24

57

517

A.I.Warper

@AIWarper

2 months

The multiverse crossovers are getting out of hand...

15

33

322

A.I.Warper

@AIWarper

2 months

How to train your dragon 🐉

19

33

316

A.I.Warper

@AIWarper

6 months

Fun experiments with #anime to live action. Had some issues with: - Blindfold color (LoRa was trained on black) - Getting 1 eye to show - Hand is a bit janky Could likely fix this with more experimentation #anime #JujustuKaisen #AIArtCommuity

22

41

294

A.I.Warper

@AIWarper

4 months

While I’m away from my PC for the week I’ll be sharing some of my old content for my new followers that may have missed it! (And to keep the algorithm from punishing me 😂) Here’s Thanos getting ready for beach season 💪

23

25

294

A.I.Warper

@AIWarper

5 months

AnimateLCM Stable Video Diffusion now working in ComfyUI thanks to Kijai! 4 step video generation...🩵 Link below 👇

12

37

275

A.I.Warper

@AIWarper

8 months

Looking for an alternative to Animate Anyone? These guys just beat them to release… code available now and initial test look promising! Link below 👇

16

33

277

A.I.Warper

@AIWarper

4 months

For everyone in my DMs asking for a tutorial 👇 Please like and share this so I don't have to reply to 5000 people 😂

VIGGLE + COMFYUI - CRAZY COMBO

Taking the power of Viggle and running it through ComfyUI!!Topaz Labs: https://topazlabs.com/ref/2377/HOW TO SUPPORT MY CHANNEL-Support me by joining my Patr...

www.youtube.com

6

48

279

A.I.Warper

@AIWarper

2 months

One of the most creative ideas I’ve seen using AI. Very well done by Enigmatic as always.

3

29

251

A.I.Warper

@AIWarper

2 months

Another absolute master class in vid2vid style transfer from Enigmatic_E The man has 3600 followers…. wtf are y’all doing man go follow him 😂

12

31

242

A.I.Warper

@AIWarper

2 months

Six months ago I attempted this shot transformation when AnimateDiff first released Still not perfect... but we've come a LONG way. I did this entire shot in less than half a day 🤯 PLEASE LIKE AND SHARE IT HELPS ME GROW 💙

30

44

242

A.I.Warper

@AIWarper

6 months

#AI as a render engine for our 3D models? 👀 Here I compiled a very crude 3D model with assets I pulled from Sketchfab. I then exported to AccuRIG for rigging and animation. #AIArtwork #Pikachu #Pokemon #3DCG #3DModel

12

26

234

A.I.Warper

@AIWarper

4 months

This is something I’ve never shared before! Had the opportunity to collaborate with Corridor Crew to develop VFX with AI to present at an annual event hosted by the HPA (Hollywood Press Association) as a demo Very fun project 🤍 I’ll share another clip from this tomorrow

12

17

232

A.I.Warper

@AIWarper

6 months

Impatient? Want to try the new Stability Cascade model? Look no further:

12

37

224

A.I.Warper

@AIWarper

4 months

Today’s share: This was my first viral video with over 8 million views and impressions! What a wild couple months since then. There were some additional comments under the original thread if you wanted more information

9

17

223

A.I.Warper

@AIWarper

2 months

Some more object removal benchmarks with ComfyUI

10

21

220

A.I.Warper

@AIWarper

2 months

Princess Peach does not simply walk into Mordor

11

14

215

A.I.Warper

@AIWarper

3 months

*Insert Toby McGuire Glasses Meme*

12

24

203

A.I.Warper

@AIWarper

1 month

Memegenerator.exe loading The LivePortrait team did say their tool works on video… I wonder if this is a sneak peek 😂

12

33

204

A.I.Warper

@AIWarper

3 months

For your viewing and listening pleasure 🎵 Live action conversion to pixel art test with 3 shots Bonus points if you can name the iconic track

22

19

198

A.I.Warper

@AIWarper

1 month

Started getting around to testing MimicMotion. Seems a decent way of using this is to stylize the first frame of your video with img2img first then use that as your ref pose Here I turned the first frame into a "zombie" (admittedly I didn't spend a ton of time on the style)

12

26

192

A.I.Warper

@AIWarper

4 months

You can use Viggle for FREE in their discord. The crazy thing is this is a 27s video and it legit took less than 5 mins to get this result 🤯 @ViggleAI

Join the Viggle Discord Server!

Viggle.ai, making any character move as you want. | 4315047 members

discord.com

8

14

185

A.I.Warper

@AIWarper

2 months

One does not simply....go this hard

9

20

192

A.I.Warper

@AIWarper

4 months

Who needs photoshop when you got ComfyUI CleanPlate generator.exe loading

10

184

A.I.Warper

@AIWarper

6 months

Vfx artists hate 'em for this one trick 🧌

12

21

178

A.I.Warper

@AIWarper

4 months

I'M JUST SATORU! Here's my test of ViggleAI + PikaLabs LipSync Volume 🆙 PLEASE LIKE AND SHARE IF YOU ENJOY THIS CONTENT 🩵

6

32

177

A.I.Warper

@AIWarper

8 months

Built my own Blender model with UVs painted to reflect the DensePose color palette Feel free to give this a try if you want something to test with! TAG ME IN YOUR CREATIONS <3 #magicanimate

5

17

175

A.I.Warper

@AIWarper

2 months

Beep Beep

7

12

176

A.I.Warper

@AIWarper

6 months

🧵Small Thread on Anime LipSync🧵 My hope in sharing some of these results is that it sparks an idea for someone to figure this out in a more efficient workflow Tools used: ComfyUI Adobe Animate 2024 Davinci Resolve 1/17

10

31

174

A.I.Warper

@AIWarper

8 months

Also….. I’m done with the matrix for now haha this will be my last one (I’m sick of looking at it) On to new things 🩵

20

1

168

A.I.Warper

@AIWarper

4 months

One of my tests from Nov 2023 I really should revisit this now that I look at it 😅

14

5

173

A.I.Warper

@AIWarper

2 months

hmmmmm.......

12

8

173

A.I.Warper

@AIWarper

3 months

"Hi I'm A.I.Warper and I make silly videos..."

23

26

170

A.I.Warper

@AIWarper

28 days

Live portrait test thread Playing with styling the source video first then using LivePortrait to bring the performance back

15

162

A.I.Warper

@AIWarper

1 month

If this works as well as I want it too... could be a game changer for vid2vid - code this week (allegedly) Link to paper in comment below to avoid nuking the algo 👇

9

24

161

A.I.Warper

@AIWarper

2 months

Open Sora model release? 🤔 anyone test yet?

GitHub - hpcaitech/Open-Sora: Open-Sora: Democratizing Efficient Video Production for All

Open-Sora: Democratizing Efficient Video Production for All - hpcaitech/Open-Sora

github.com

9

26

162

A.I.Warper

@AIWarper

8 months

Stable Video Diffusion Temporal ControlNET

GitHub - CiaraStrawberry/svd-temporal-controlnet

Contribute to CiaraStrawberry/svd-temporal-controlnet development by creating an account on GitHub.

github.com

4

19

158

A.I.Warper

@AIWarper

1 month

Retro anime with lip sync and eye direction Ignore the hand. Could fix it with masking but I am lazy 🤣

13

17

158

A.I.Warper

@AIWarper

4 months

Firstly, I rotoscoped the subject using After effects. This isn’t necessary but I needed a mask for a later step anyways so - why not right?

5

14

154

A.I.Warper

@AIWarper

8 days

Segment anything 2 release!

GitHub - facebookresearch/segment-anything-2: The repository provides code for running inference...

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use th...

github.com

3

17

146

A.I.Warper

@AIWarper

8 months

ImageDream img to 3D Looks like code available 🩵

GitHub - bytedance/ImageDream: The code releasing for https://image-dream.github.io/

The code releasing for https://image-dream.github.io/ - bytedance/ImageDream

github.com

2

11

141

A.I.Warper

@AIWarper

5 months

Transparent image generation with SDXL Looks like it can do glass and other transparent materials etc… pretty cool!

6

19

145

A.I.Warper

@AIWarper

2 months

Image 1: SD text 2 img Image 2: Rodin image to 3D Model Image 3: Rigged 3D model in Blender Image 4: My awful keyframed idle pose pixel art animation (stylized to pixel w/ Comfy) Proof of concept for prototyping game assets. Someone experienced could really work fast like this.

11

29

144

A.I.Warper

@AIWarper

4 months

Baby has arrived! Mom is doing very well and everything went smooth 💙

25

1

138

A.I.Warper

@AIWarper

4 months

I’ve gotten so many requests for AI in the past 48hrs it’s staggering… Someone should you know.. start a company to provide this service. I can provide the workflows and workforce of people through my connections I just don’t want to manage a company 😂

31

4

136

A.I.Warper

@AIWarper

4 days

FLUX --> LivePortrait

7

17

227

A.I.Warper

@AIWarper

4 months

Of all the reactions I've had to my content this one is oddly the most surreal to me. I started WoW 3 months before BC released, had 5 Cutting Edge kills, and spent thousands of hours in Azeroth Weird feeling having a video go this viral out of no where

12

2

133

A.I.Warper

@AIWarper

2 months

FPS is way too high, but pretty cool vintage 90s anime vibes

4

7

135

A.I.Warper

@AIWarper

8 months

Thank you for the overwhelming support on this post. Just a heads up I’m not overly checking the comments or the quoted retweets as the majority of them have just been doomers trying to kill my morale If you quoted with positive words thank you so much (I’m not ignoring you)🩵

A.I.Warper

@AIWarper

8 months

Here's probably one of the hardest shots I've ever attempted 1) Two characters 2) Two different race of humans (prompt bleeding) 3) Very fast paced with a lot of motion Was very fun and I learned a lot. Also develop a bunch of new workflow #StableDiffusion #Matrix #Pixar

1K

521

5K

45

1

135

A.I.Warper

@AIWarper

1 month

Well that’s cool Universal CN for SDXL. Appears to work in ComfyUI with Kosinkadink’s CN nodes and adapts to whatever you feed it (depth, normals, etc)

xinsir/controlnet-union-sdxl-1.0 · Hugging Face

huggingface.co

5

22

130

A.I.Warper

@AIWarper

6 months

Lip flap testing Lip shapes generated in Comfy UI then auto lipsynced to audio Very promising so far... Will post something longer with audio later

12

23

126

A.I.Warper

@AIWarper

3 months

And this is why I don't share my workflows 😉

KREA AI

@krea_ai

3 months

Krea Video is here! this is how it works 👇 - (sound on)

4K

439

3K

11

4

126

A.I.Warper

@AIWarper

6 months

So...... is #gemini trolling me? Can't make this stuff up.... cannot believe it replied with this lol

14

4

127

A.I.Warper

@AIWarper

2 months

Image 2 video example with LUMA AI This is an example of taking one of their outputs and cleaning it up / fixing the camera angle How good is this? 🥵 #lumadreammachine

10

15

125

A.I.Warper

@AIWarper

21 days

My timeline is rapidly being engulfed by politics / click bait..... I just want to see AI stuff plz thanks

27

6

123

A.I.Warper

@AIWarper

8 months

Today's Experiments testing out #MagicAnimate What do you guys think? I am pretty happy with the results I built a custom rig in Blender to output the Densepose Animation #AIart #AIArtCommuity #AnimateDiff #Anime #yugioh

9

11

119

A.I.Warper

@AIWarper

8 months

Longer context window for AnimateDiff? Yes please ✅

GitHub - Lightricks/LongAnimateDiff

Contribute to Lightricks/LongAnimateDiff development by creating an account on GitHub.

github.com

5

21

120

A.I.Warper

@AIWarper

4 months

A note to my community as I cross 10k followers: Thank you so much! I never dreamed 10,000 people would be interested in my knob turning experiments 🤍💙 An update on life: My wife is due to give birth literally any day now so if I go dark for a bit - you know why 👶🍼

22

2

119

A.I.Warper

@AIWarper

6 months

Some random LCM results: Checkpoints used starting from top (left to right) 1: Source video 2: Disney Pixar Cartoon type B (10/25 steps) 3: Arthemy Comics 4: Flat-2D Animerge 5: IcoMix (LCM Lora was 1.5 strength) 6: Same as 2, but empty latents and princess peach prompt

11

121

A.I.Warper

@AIWarper

1 month

On this new season eve it feels appropriate to reshare this 🧸 clip. Let’s pray season 3 of The Bear gives us another masterpiece of any episode as good as Forks!

11

16

121

A.I.Warper

@AIWarper

5 months

New anime upscale now available in ComfyUI Very fast and the results are probably the best I’ve seen

GitHub - kijai/ComfyUI-APISR-KJ: Node to use APISR upscale models in ComfyUI

Node to use APISR upscale models in ComfyUI. Contribute to kijai/ComfyUI-APISR-KJ development by creating an account on GitHub.

github.com

4

22

120

A.I.Warper

@AIWarper

5 months

More fun with motion loras! Very addicting to play with

6

9

113

A.I.Warper

@AIWarper

2 months

Not a huge style shift, but look how well it preserves expression and lip movement. Basically 1:1 with the original On a side note..... A traditional Western comic style is proving to be incredibly difficult to produce Looking for heavy black inked lines and shading

8

14

116

A.I.Warper

@AIWarper

5 months

One does not simply use Reverse Noise

7

5

114

A.I.Warper

@AIWarper

14 days

Viggle pipeline still the goat For 2D flat stuff like this it's significantly better than mimic and LivePortrait

6

18

114

A.I.Warper

@AIWarper

4 months

It gave me this video. The next steps involved compositing back into the source footage. This was the step that I need to improve on and have since come up with a better method for doing it involving in painting and gen fill