Loading video...

Video Failed to Load

Go Home

comfyUI-LongLook: consistent motion for Wan 2.2; - uses FreeLong Spectral Blending to smooth motion reversal in 81+ frame generations; combines global & local attention via FFT to enforce motion prompts, especially for i2v.

wildmindai's profile picture

Wildminder

9,287 subscribers

12,234 views • 9 months ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

RobotMDM, by Disney Research, combines diffusion-based motion generation with RL to produce physics-aware humanoid motions from text prompts. Trained on human motion data with a reward surrogate for physical feasibility, it ensures realistic motions.
0:56

RobotMDM, by Disney Research, combines diffusion-based motion generation with RL to produce physics-aware humanoid motions from text prompts. Trained on human motion data with a reward surrogate for physical feasibility, it ensures realistic motions.

The Humanoid Hub

22,943 views • 2 years ago

Your creative workflow is missing these tools. - Motion templates for creatives - Turn your images into motion in minutes. - A new way to create images - Dithered icons. In motion. - Discover polished AI prompts and visual references
0:20

Your creative workflow is missing these tools. - Motion templates for creatives - Turn your images into motion in minutes. - A new way to create images - Dithered icons. In motion. - Discover polished AI prompts and visual references

Simoom

16,698 views • 7 days ago

Wan 2.5-Preview: Know You Better. Create Freely. Now live—built to follow complex instructions, support dynamic motion and structured prompts, and generate smooth 10s videos in 1080P at 24fps. Your vision, perfectly executed. Stay tuned with us for more Wan 2.5 insights!
0:53

Wan 2.5-Preview: Know You Better. Create Freely. Now live—built to follow complex instructions, support dynamic motion and structured prompts, and generate smooth 10s videos in 1080P at 24fps. Your vision, perfectly executed. Stay tuned with us for more Wan 2.5 insights!

Alibaba Cloud

1,113,795 views • 10 months ago

Face Replacement and De-aging Demo: VFX artist Doug Hogan built a fully automated face swap workflow inside ComfyUI. Florence 2 → SAM2 → WAN Video. One pipeline. Temporally consistent across every frame. Not a pixel swap. The workflow detects the head, segments it, drives motion from the original performance, and uses a reference image + auto prompt to transfer the identity while preserving lighting and timing. Face replacement. De-aging. All inside ComfyUI. Click the link below for the workflow 👇
4:13

Face Replacement and De-aging Demo: VFX artist Doug Hogan built a fully automated face swap workflow inside ComfyUI. Florence 2 → SAM2 → WAN Video. One pipeline. Temporally consistent across every frame. Not a pixel swap. The workflow detects the head, segments it, drives motion from the original performance, and uses a reference image + auto prompt to transfer the identity while preserving lighting and timing. Face replacement. De-aging. All inside ComfyUI. Click the link below for the workflow 👇

ComfyUI

14,877 views • 3 months ago

Wan 3.0 is live on OpenArt 🔊 • Native 30-second generation in a single pass — no stitching • Consistent motion, lighting & subjects throughout • 480p, 720p & 1080p outputs • Upgrade to Pro or Wonder for unlimited generations this week (Aug 24–31). Available now on OpenArt:
0:25

Wan 3.0 is live on OpenArt 🔊 • Native 30-second generation in a single pass — no stitching • Consistent motion, lighting & subjects throughout • 480p, 720p & 1080p outputs • Upgrade to Pro or Wonder for unlimited generations this week (Aug 24–31). Available now on OpenArt:

OpenArt

21,242 views • 1 month ago

🔥 Introducing MVLift: Generate realistic 3D motion without any 3D training data - just using 2D poses from monocular videos! Applicable to human motion, human-object interaction & animal motion. Joint work w/ Jiajun Wu & Karen 💡 How? We reformulate 3D motion estimation as generating consistent multi-view 2D pose sequences. Our framework uses 2D motion diffusion to progressively establish multi-view consistency, requiring only single-view 2D pose sequences for training. Project: Video with demonstration: Paper:
0:21

🔥 Introducing MVLift: Generate realistic 3D motion without any 3D training data - just using 2D poses from monocular videos! Applicable to human motion, human-object interaction & animal motion. Joint work w/ Jiajun Wu & Karen 💡 How? We reformulate 3D motion estimation as generating consistent multi-view 2D pose sequences. Our framework uses 2D motion diffusion to progressively establish multi-view consistency, requiring only single-view 2D pose sequences for training. Project: Video with demonstration: Paper:

Jiaman Li

15,886 views • 1 year ago

Generating flipbooks in ComfyUI. Nothing too crazy, but these little AI workflows are definitely helpful! 🐦 When I input a black frame to the FirstLastFrame node I got some decent impulse textures. using Wan 2.2 btw. For the bird to work i had to generate between 30-60 frames and extract every 4 frames to get better wing motion. In UEFN / Unreal Engine i'm offsetting the animation phase for every particle to get a more random result.
0:59

Generating flipbooks in ComfyUI. Nothing too crazy, but these little AI workflows are definitely helpful! 🐦 When I input a black frame to the FirstLastFrame node I got some decent impulse textures. using Wan 2.2 btw. For the bird to work i had to generate between 30-60 frames and extract every 4 frames to get better wing motion. In UEFN / Unreal Engine i'm offsetting the animation phase for every particle to get a more random result.

Jerome | InsaneUnreal

10,952 views • 8 months ago

From an instant thought to a 30-second cinematic sequence in crisp 1080p HD. Wan Alibaba Cloud Wan 3.0 has officially landed on SocialSight! Keep your characters consistent across every frame, direct motion like a pro, and generate the moment before it passes.
0:29

From an instant thought to a 30-second cinematic sequence in crisp 1080p HD. Wan Alibaba Cloud Wan 3.0 has officially landed on SocialSight! Keep your characters consistent across every frame, direct motion like a pro, and generate the moment before it passes.

SocialSight

117,064 views • 1 month ago

Video2Humanoid. Still trouble with bad retargeted humanoid motions? Humanoids now are Easy to track any video using our new Neural Motion Retargeting (NMR) Method. Optimization-based motion retargeting methods like IK and GMR are solving a non-convex problem frame by frame, which makes them sensitive to initialization, hard to tune, and prone to poor local minima. The result is familiar: joint discontinuities, self-collisions, and unstable foot contact. Our solution is simple but effective : instead of optimizing each frame independently, learn the mapping between human motion and robot motion as a distribution via networks. 📊Results on Unitree G1 - 0 joint jumps - 54% fewer self-collision frames - 61% fewer joint limit violations - Faster convergence for downstream control policy training NMR turns motion retargeting from a fragile optimization problem into a learned, scalable pipeline for more stable humanoid motion. please visit for more visualization results. opensource:
0:54

Video2Humanoid. Still trouble with bad retargeted humanoid motions? Humanoids now are Easy to track any video using our new Neural Motion Retargeting (NMR) Method. Optimization-based motion retargeting methods like IK and GMR are solving a non-convex problem frame by frame, which makes them sensitive to initialization, hard to tune, and prone to poor local minima. The result is familiar: joint discontinuities, self-collisions, and unstable foot contact. Our solution is simple but effective : instead of optimizing each frame independently, learn the mapping between human motion and robot motion as a distribution via networks. 📊Results on Unitree G1 - 0 joint jumps - 54% fewer self-collision frames - 61% fewer joint limit violations - Faster convergence for downstream control policy training NMR turns motion retargeting from a fragile optimization problem into a learned, scalable pipeline for more stable humanoid motion. please visit for more visualization results. opensource:

Xiao-Xiao Long

24,245 views • 5 months ago

The Labour Party has been forced to allow an emergency motion on Palestine today. We urge all delegates to stand up for the truth that Israel is committing genocide & vote for motion "Peace in the Middle East 2".
0:59

The Labour Party has been forced to allow an emergency motion on Palestine today. We urge all delegates to stand up for the truth that Israel is committing genocide & vote for motion "Peace in the Middle East 2".

Palestine Solidarity Campaign

23,394 views • 1 year ago

In order to consider the question "is this motion legal?", I took a frame by frame look at the 49ers motion plays against the Rams My main takeaway was that the whole operation deserves a lot of credit for their timing, including Purdy's cadence and Juszczyk's patience
9:23

In order to consider the question "is this motion legal?", I took a frame by frame look at the 49ers motion plays against the Rams My main takeaway was that the whole operation deserves a lot of credit for their timing, including Purdy's cadence and Juszczyk's patience

Shawn Syed

89,283 views • 25 days ago

Image. Then motion. GPT Image 2 + Seedance 2.0 are now both live on Pollo AI. GPT Image 2 → structured layouts, sharp text, consistent characters — ready for ads, products, and social content. Seedance 2.0 → cinematic, stable, smooth video — straight from your image. From image to motion, the workflow is fast, seamless, and powerful.
0:20

Image. Then motion. GPT Image 2 + Seedance 2.0 are now both live on Pollo AI. GPT Image 2 → structured layouts, sharp text, consistent characters — ready for ads, products, and social content. Seedance 2.0 → cinematic, stable, smooth video — straight from your image. From image to motion, the workflow is fast, seamless, and powerful.

Pollo AI

432,024 views • 5 months ago

Tracking Everything Everywhere All at Once paper page: present a new test-time optimization method for estimating dense and long-range motion from a video sequence. Prior optical flow or particle video tracking algorithms typically operate within limited temporal windows, struggling to track through occlusions and maintain global consistency of estimated motion trajectories. We propose a complete and globally consistent motion representation, dubbed OmniMotion, that allows for accurate, full-length motion estimation of every pixel in a video. OmniMotion represents a video using a quasi-3D canonical volume and performs pixel-wise tracking via bijections between local and canonical space. This representation allows us to ensure global consistency, track through occlusions, and model any combination of camera and object motion. Extensive evaluations on the TAP-Vid benchmark and real-world footage show that our approach outperforms prior state-of-the-art methods by a large margin both quantitatively and qualitatively.
5:18

Tracking Everything Everywhere All at Once paper page: present a new test-time optimization method for estimating dense and long-range motion from a video sequence. Prior optical flow or particle video tracking algorithms typically operate within limited temporal windows, struggling to track through occlusions and maintain global consistency of estimated motion trajectories. We propose a complete and globally consistent motion representation, dubbed OmniMotion, that allows for accurate, full-length motion estimation of every pixel in a video. OmniMotion represents a video using a quasi-3D canonical volume and performs pixel-wise tracking via bijections between local and canonical space. This representation allows us to ensure global consistency, track through occlusions, and model any combination of camera and object motion. Extensive evaluations on the TAP-Vid benchmark and real-world footage show that our approach outperforms prior state-of-the-art methods by a large margin both quantitatively and qualitatively.

AK

280,570 views • 3 years ago

Introducing Wan 2.5 🚀 Create immersive AI videos with fluid motion, built-in audio and voice. Supporting text-to-video and start images Available now for all users
0:23

Introducing Wan 2.5 🚀 Create immersive AI videos with fluid motion, built-in audio and voice. Supporting text-to-video and start images Available now for all users

Magnific

170,229 views • 1 year ago

Before the Void Swallows You Made with Wan 3.0 30s in one go, the style stays consistent from start to finish, and the motion and overall coherence feel noticeably better. Also just noticed WAN 3.0 is ranked first on Artificial Analysis for both Video Editing (with Audio) and Text-to-Video (with Audio).
0:30

Before the Void Swallows You Made with Wan 3.0 30s in one go, the style stays consistent from start to finish, and the motion and overall coherence feel noticeably better. Also just noticed WAN 3.0 is ranked first on Artificial Analysis for both Video Editing (with Audio) and Text-to-Video (with Audio).

Latte

67,581 views • 1 month ago

Runway Gen 4.5 has two new features. Motion Sketch and Character Swap are now in-built apps for the tool. Sketch camera motion with annotations on a start frame to control the movement. Swap a character with two images before the video step. Here's how!
0:29

Runway Gen 4.5 has two new features. Motion Sketch and Character Swap are now in-built apps for the tool. Sketch camera motion with annotations on a start frame to control the movement. Swap a character with two images before the video step. Here's how!

Jerrod Lew

13,233 views • 8 months ago

Wan2.2 is now natively supported in ComfyUI on Day 0! 🔹 A next-gen video model with MoE (Mixture of Experts) architecture with dual noise experts, under Apache 2.0 license! - Cinematic-level Aesthetic Control - Large-scale Complex Motion - Precise Semantic Compliance 📚 Versions available: - Wan2.2-TI2V-5B: FP16 - Wan2.2-I2V-14B: FP16/FP8 - Wan2.2-T2V-14B: FP16/FP8 💻 Down to 8GB VRAM requirement for the 5B version with ComfyUI auto-offloading.
0:55

Wan2.2 is now natively supported in ComfyUI on Day 0! 🔹 A next-gen video model with MoE (Mixture of Experts) architecture with dual noise experts, under Apache 2.0 license! - Cinematic-level Aesthetic Control - Large-scale Complex Motion - Precise Semantic Compliance 📚 Versions available: - Wan2.2-TI2V-5B: FP16 - Wan2.2-I2V-14B: FP16/FP8 - Wan2.2-T2V-14B: FP16/FP8 💻 Down to 8GB VRAM requirement for the 5B version with ComfyUI auto-offloading.

ComfyUI

84,618 views • 1 year ago

💿NEW for @vorihq AI video gen is getting quite good... But the best way to get precise control over every frame is still to control every single frame So for Afterimage’s latest, we built miniature sets and filmed everything in stop motion
0:47

💿NEW for @vorihq AI video gen is getting quite good... But the best way to get precise control over every frame is still to control every single frame So for Afterimage’s latest, we built miniature sets and filmed everything in stop motion

josh

18,364 views • 5 months ago