正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

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.

Wildminder

9,287 subscribers

12,234 次观看 • 7 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

🚀 Just released a LoRA for Wan 2.1 that adds realistic drone-style push-in motion. Model: Wan 2.1 I2V - 14B 720p Trained on 100 clips — and refined over 40+ versions. Trigger: Push-in_camera 🎥 + ComfyUI workflow included for easy use Perfect if you want your videos to actually *move*. 👉 #AI #LoRA #wan21 #generativevideo ComfyUI Made in collaboration with Kartel.ai

🚀 Just released a LoRA for Wan 2.1 that adds realistic drone-style push-in motion. Model: Wan 2.1 I2V - 14B 720p Trained on 100 clips — and refined over 40+ versions. Trigger: Push-in_camera 🎥 + ComfyUI workflow included for easy use Perfect if you want your videos to actually move. 👉 #AI #LoRA #wan21 #generativevideo ComfyUI Made in collaboration with Kartel.ai

Lovis Odin

18,366 次观看 • 1 年前

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.

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 次观看 • 1 年前

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!

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 次观看 • 7 个月前

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 👇

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

13,948 次观看 • 1 个月前

🔥 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:

🔥 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 次观看 • 1 年前

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.

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 次观看 • 6 个月前

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:

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

24,193 次观看 • 3 个月前

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".

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

23,268 次观看 • 10 个月前

Introducing Move Pro 2.0: the next evolution in high-precision motion capture. Local and secure processing, faster performance with dedicated GPUs, cloud scaling options, and second-generation AI models for enhanced accuracy. Perfect for capturing motion from small studios to large stadiums.

Introducing Move Pro 2.0: the next evolution in high-precision motion capture. Local and secure processing, faster performance with dedicated GPUs, cloud scaling options, and second-generation AI models for enhanced accuracy. Perfect for capturing motion from small studios to large stadiums.

Move AI

14,141 次观看 • 1 年前

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.

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

431,569 次观看 • 3 个月前

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.

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

280,547 次观看 • 3 年前

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

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 次观看 • 10 个月前

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!

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

13,177 次观看 • 6 个月前

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.

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

84,496 次观看 • 1 年前

💿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

💿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 次观看 • 3 个月前

How to CREATE Long AI Videos with Consistent Characters in (2026) Learn how to create long cinematic AI videos with consistent characters, environments, and motion. Learn how build reference assets, chain clips together with previous video generations, and structure prompts so the final result feels like one continuous film instead of disconnected AI shots. This information is for both the new and existing Youtube creators, who want to learn how to create good videos, with consistent characters. Watch to the end, to learn properly. Like, Share and Retweet. Follow BIG CHRIS

How to CREATE Long AI Videos with Consistent Characters in (2026) Learn how to create long cinematic AI videos with consistent characters, environments, and motion. Learn how build reference assets, chain clips together with previous video generations, and structure prompts so the final result feels like one continuous film instead of disconnected AI shots. This information is for both the new and existing Youtube creators, who want to learn how to create good videos, with consistent characters. Watch to the end, to learn properly. Like, Share and Retweet. Follow BIG CHRIS

BIG CHRIS

46,409 次观看 • 2 个月前

How was debate shut down? Lucas North proposed a motion to move to “next business” & ask the Liberal Democrats not to discuss Liberal Voice for Women’s motion arguing for women’s fair representation in the party Watch Zoe Hollowood 🔶 🇪🇺🇺🇦’s speech Lucas won the vote Two thirds majority - (3 clips)

How was debate shut down? Lucas North proposed a motion to move to “next business” & ask the Liberal Democrats not to discuss Liberal Voice for Women’s motion arguing for women’s fair representation in the party Watch Zoe Hollowood 🔶 🇪🇺🇺🇦’s speech Lucas won the vote Two thirds majority - (3 clips)

teresa smith

118,225 次观看 • 10 个月前

Precision in motion. ✨ Watch the new BEASTBOX GODSPEED transition from beast to bot in one fluid sequence. ⚡️Every movement is engineered for smooth, satisfying flow. 🐎 This is speed, refined.

Precision in motion. ✨ Watch the new BEASTBOX GODSPEED transition from beast to bot in one fluid sequence. ⚡️Every movement is engineered for smooth, satisfying flow. 🐎 This is speed, refined.

52TOYS BEASTBOX

15,831 次观看 • 6 个月前

Lil Durk Update. Drew Findling working every angle. This time it’s a motion to recuse judges in the district and prosecutors in the local USAO office. Worked for Thug. Will it work for Durk?

Lil Durk Update. Drew Findling working every angle. This time it’s a motion to recuse judges in the district and prosecutors in the local USAO office. Worked for Thug. Will it work for Durk?

Mohammed (Moe) Gangat

75,815 次观看 • 8 个月前