Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

KEEP can enhance video face super-resolution by maintaining consistency across frames! It uses Kalman filtering to improve facial details, working well on both synthetic and real-world videos. Links ⬇️

dreamingtulpa's profile picture

Dreaming Tulpa 🥓👑

55,198 subscribers

252,813 Aufrufe • vor 2 Jahren •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution paper page: Text-based diffusion models have exhibited remarkable success in generation and editing, showing great promise for enhancing visual content with their generative prior. However, applying these models to video super-resolution remains challenging due to the high demands for output fidelity and temporal consistency, which is complicated by the inherent randomness in diffusion models. Our study introduces Upscale-A-Video, a text-guided latent diffusion framework for video upscaling. This framework ensures temporal coherence through two key mechanisms: locally, it integrates temporal layers into U-Net and VAE-Decoder, maintaining consistency within short sequences; globally, without training, a flow-guided recurrent latent propagation module is introduced to enhance overall video stability by propagating and fusing latent across the entire sequences. Thanks to the diffusion paradigm, our model also offers greater flexibility by allowing text prompts to guide texture creation and adjustable noise levels to balance restoration and generation, enabling a trade-off between fidelity and quality. Extensive experiments show that Upscale-A-Video surpasses existing methods in both synthetic and real-world benchmarks, as well as in AI-generated videos, showcasing impressive visual realism and temporal consistency.
1:45

Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution paper page: Text-based diffusion models have exhibited remarkable success in generation and editing, showing great promise for enhancing visual content with their generative prior. However, applying these models to video super-resolution remains challenging due to the high demands for output fidelity and temporal consistency, which is complicated by the inherent randomness in diffusion models. Our study introduces Upscale-A-Video, a text-guided latent diffusion framework for video upscaling. This framework ensures temporal coherence through two key mechanisms: locally, it integrates temporal layers into U-Net and VAE-Decoder, maintaining consistency within short sequences; globally, without training, a flow-guided recurrent latent propagation module is introduced to enhance overall video stability by propagating and fusing latent across the entire sequences. Thanks to the diffusion paradigm, our model also offers greater flexibility by allowing text prompts to guide texture creation and adjustable noise levels to balance restoration and generation, enabling a trade-off between fidelity and quality. Extensive experiments show that Upscale-A-Video surpasses existing methods in both synthetic and real-world benchmarks, as well as in AI-generated videos, showcasing impressive visual realism and temporal consistency.

AK

32,849 Aufrufe • vor 2 Jahren

Marlon was shocked by the new videos that MrBeast is currently working on 😳🔥 VIDEO #1 - 100 wannabe astronauts compete in games to win a chance to go to space. VIDEO #2 - someone drives a lambo across the world to keep it (could take 4 months)
0:46

Marlon was shocked by the new videos that MrBeast is currently working on 😳🔥 VIDEO #1 - 100 wannabe astronauts compete in games to win a chance to go to space. VIDEO #2 - someone drives a lambo across the world to keep it (could take 4 months)

ryan 🤿

667,318 Aufrufe • vor 6 Monaten

A humanoid robot policy trained solely on synthetic data generated by a world model. Research Scientist Joel Jang presents NVIDIA's DreamGen pipeline: ⦿ Post-train the world model Cosmos-Predict2 with a small set of real teleoperation demos. ⦿ Prompt the world model to generate synthetic video data with verbs and scenarios not used in the world model’s post-training. ⦿ Auto-label synthetic video data with action sequences. ⦿ Train robot policies using only synthetic data. That's it. Deploy zero-shot to a real humanoid robot.
2:05

A humanoid robot policy trained solely on synthetic data generated by a world model. Research Scientist Joel Jang presents NVIDIA's DreamGen pipeline: ⦿ Post-train the world model Cosmos-Predict2 with a small set of real teleoperation demos. ⦿ Prompt the world model to generate synthetic video data with verbs and scenarios not used in the world model’s post-training. ⦿ Auto-label synthetic video data with action sequences. ⦿ Train robot policies using only synthetic data. That's it. Deploy zero-shot to a real humanoid robot.

The Humanoid Hub

20,968 Aufrufe • vor 1 Jahr

🚀🚀🚀Introducing HY World 1.5 (WorldPlay)! We have now open-sourced the most systemized, comprehensive real-time world model framework in the industry. In HY World 1.5, we develop WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off between speed and memory that limits current methods. You can generate and explore 3D worlds simply by inputting text or images. Walk, look around, and interact like you're playing a game. Highlights: 🔹Real-Time: Generates long-horizon streaming video at 24 FPS with superior consistency. 🔹Geometric Consistency: Achieved using a Reconstituted Context Memory mechanism to dynamically rebuild context from past frames to alleviate memory attenuation 🔹Robust Control: Uses a Dual Action Representation for robust response to user keyboard and mouse inputs. 🔹Versatile Applications: Supports both first-person and third-person perspectives, enabling applications like promptable events and infinite world extension. 👉🏻Try it now: 🌐Project Page: 🔗Github: 🤗Hugging Face: 📄Technical Report:
1:06

🚀🚀🚀Introducing HY World 1.5 (WorldPlay)! We have now open-sourced the most systemized, comprehensive real-time world model framework in the industry. In HY World 1.5, we develop WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off between speed and memory that limits current methods. You can generate and explore 3D worlds simply by inputting text or images. Walk, look around, and interact like you're playing a game. Highlights: 🔹Real-Time: Generates long-horizon streaming video at 24 FPS with superior consistency. 🔹Geometric Consistency: Achieved using a Reconstituted Context Memory mechanism to dynamically rebuild context from past frames to alleviate memory attenuation 🔹Robust Control: Uses a Dual Action Representation for robust response to user keyboard and mouse inputs. 🔹Versatile Applications: Supports both first-person and third-person perspectives, enabling applications like promptable events and infinite world extension. 👉🏻Try it now: 🌐Project Page: 🔗Github: 🤗Hugging Face: 📄Technical Report:

Tencent Hy

189,802 Aufrufe • vor 8 Monaten

Dear #SidraChain family Now facial recognition and ST token claim is working perfectly just wait a minute after face detection You can see the below video for details and if the error is still appears try to clear catch of the app or reinstall it Keep contributing to #SidraBank blockchain 👍 #SidraBankNews #Sidrachain #SidraGem #SidraFamily #sidra #SidraIsComing
0:35

Dear #SidraChain family Now facial recognition and ST token claim is working perfectly just wait a minute after face detection You can see the below video for details and if the error is still appears try to clear catch of the app or reinstall it Keep contributing to #SidraBank blockchain 👍 #SidraBankNews #Sidrachain #SidraGem #SidraFamily #sidra #SidraIsComing

SidraBank News

13,384 Aufrufe • vor 6 Monaten

Policies trained on real robot data via imitation can be surprisingly capable. But for domains like dexterous manipulation, they are often not quite good enough: they move slowly, miss grasps, make unreliable contact, and fail under small perturbations. Can we improve them without any additional data collection on the real robot? In SCORE, we show that we can improve real-world diffusion/flow policies cheaply by using simulation to simply learn how to steer them on deployment. This leads to large gains in real-world success and speed across a variety of tasks, without requiring additional real-world experience: 🧵 (1/10)
0:44

Policies trained on real robot data via imitation can be surprisingly capable. But for domains like dexterous manipulation, they are often not quite good enough: they move slowly, miss grasps, make unreliable contact, and fail under small perturbations. Can we improve them without any additional data collection on the real robot? In SCORE, we show that we can improve real-world diffusion/flow policies cheaply by using simulation to simply learn how to steer them on deployment. This leads to large gains in real-world success and speed across a variety of tasks, without requiring additional real-world experience: 🧵 (1/10)

Abhishek Gupta

34,775 Aufrufe • vor 1 Monat

Australia and New Zealand are family. In a more uncertain world, our bond matters more than ever. By working together, we can create jobs and grow both our economies. We’re making it easier for businesses to invest and operate on both sides of the Tasman. And we’ll continue working closely on security and defence. When one of our countries does well, we both do. And together, we’re building on that success.
0:23

Australia and New Zealand are family. In a more uncertain world, our bond matters more than ever. By working together, we can create jobs and grow both our economies. We’re making it easier for businesses to invest and operate on both sides of the Tasman. And we’ll continue working closely on security and defence. When one of our countries does well, we both do. And together, we’re building on that success.

Anthony Albanese

77,204 Aufrufe • vor 1 Jahr

UEDCL is committed to implementing reliability improvement projects across our communities. We call upon all community members to support and embrace these initiatives, as they are designed to enhance service delivery and improve the quality of life for everyone. By working together we can build a stronger, more reliable system that benefits us all. For more videos🎥🎞️, visit our ▶️youtube channel: #UEDCLinService #LightingUpYourWorld
2:04

UEDCL is committed to implementing reliability improvement projects across our communities. We call upon all community members to support and embrace these initiatives, as they are designed to enhance service delivery and improve the quality of life for everyone. By working together we can build a stronger, more reliable system that benefits us all. For more videos🎥🎞️, visit our ▶️youtube channel: #UEDCLinService #LightingUpYourWorld

Uganda Electricity Distribution Company Ltd

10,645 Aufrufe • vor 1 Monat

zeroscope_v2 XL, A watermark-free Modelscope-based video model capable of generating high quality video at 1024 x 576 Model on Hugging Face : This model was trained with offset noise using 9,923 clips and 29,769 tagged frames at 24 frames, 1024x576 resolution. zeroscope_v2_XL is specifically designed for upscaling content made with zeroscope_v2_576w using vid2vid in the 1111 text2video extension by kabachuha. Leveraging this model as an upscaler allows for superior overall compositions at higher resolutions, permitting faster exploration in 576x320 (or 448x256) before transitioning to a high-resolution render. zeroscope_v2_XL uses 15.3gb of vram when rendering 30 frames at 1024x576
0:35

zeroscope_v2 XL, A watermark-free Modelscope-based video model capable of generating high quality video at 1024 x 576 Model on Hugging Face : This model was trained with offset noise using 9,923 clips and 29,769 tagged frames at 24 frames, 1024x576 resolution. zeroscope_v2_XL is specifically designed for upscaling content made with zeroscope_v2_576w using vid2vid in the 1111 text2video extension by kabachuha. Leveraging this model as an upscaler allows for superior overall compositions at higher resolutions, permitting faster exploration in 576x320 (or 448x256) before transitioning to a high-resolution render. zeroscope_v2_XL uses 15.3gb of vram when rendering 30 frames at 1024x576

AK

714,363 Aufrufe • vor 3 Jahren

Why do cars keep bobbing up and down on Indian highways? You can clearly notice it from the way the camera shakes in both videos! Even on one of the biggest expressways the Amritsar–Jamnagar Highway the ride feels bumpy. Honestly, it seems like we still haven’t mastered maintaining proper straightness and leveling. You don’t usually see this kind of unevenness on six-lane highways in many other countries. We need to admit that we still have a lot to improve when it comes to highway engineering!
0:43

Why do cars keep bobbing up and down on Indian highways? You can clearly notice it from the way the camera shakes in both videos! Even on one of the biggest expressways the Amritsar–Jamnagar Highway the ride feels bumpy. Honestly, it seems like we still haven’t mastered maintaining proper straightness and leveling. You don’t usually see this kind of unevenness on six-lane highways in many other countries. We need to admit that we still have a lot to improve when it comes to highway engineering!

Rattan Dhillon

152,318 Aufrufe • vor 5 Monaten

PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking paper page: introduce PointOdyssey, a large-scale synthetic dataset, and data generation framework, for the training and evaluation of long-term fine-grained tracking algorithms. Our goal is to advance the state-of-the-art by placing emphasis on long videos with naturalistic motion. Toward the goal of naturalism, we animate deformable characters using real-world motion capture data, we build 3D scenes to match the motion capture environments, and we render camera viewpoints using trajectories mined via structure-from-motion on real videos. We create combinatorial diversity by randomizing character appearance, motion profiles, materials, lighting, 3D assets, and atmospheric effects. Our dataset currently includes 104 videos, averaging 2,000 frames long, with orders of magnitude more correspondence annotations than prior work. We show that existing methods can be trained from scratch in our dataset and outperform the published variants. Finally, we introduce modifications to the PIPs point tracking method, greatly widening its temporal receptive field, which improves its performance on PointOdyssey as well as on two real-world benchmarks.
0:33

PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking paper page: introduce PointOdyssey, a large-scale synthetic dataset, and data generation framework, for the training and evaluation of long-term fine-grained tracking algorithms. Our goal is to advance the state-of-the-art by placing emphasis on long videos with naturalistic motion. Toward the goal of naturalism, we animate deformable characters using real-world motion capture data, we build 3D scenes to match the motion capture environments, and we render camera viewpoints using trajectories mined via structure-from-motion on real videos. We create combinatorial diversity by randomizing character appearance, motion profiles, materials, lighting, 3D assets, and atmospheric effects. Our dataset currently includes 104 videos, averaging 2,000 frames long, with orders of magnitude more correspondence annotations than prior work. We show that existing methods can be trained from scratch in our dataset and outperform the published variants. Finally, we introduce modifications to the PIPs point tracking method, greatly widening its temporal receptive field, which improves its performance on PointOdyssey as well as on two real-world benchmarks.

AK

122,567 Aufrufe • vor 3 Jahren

🚨 Awaab's Law consultation ends soon. I've made an animated video covering all details. Can you lend a hand by sharing this thread of videos & links? Critical for social housing tenants to know about the consultation and engage with it if they can. (1/3🧵)
2:06

🚨 Awaab's Law consultation ends soon. I've made an animated video covering all details. Can you lend a hand by sharing this thread of videos & links? Critical for social housing tenants to know about the consultation and engage with it if they can. (1/3🧵)

Christian Weaver

18,967 Aufrufe • vor 2 Jahren

Face swapping just leveled up. Akool Face Swap just got a major upgrade — designed to make AI video creation faster, smarter, and way more powerful. Now you can process up to 50 files in batch, swap faces directly on GIFs, and benefit from smarter face detection that automatically removes duplicate faces across uploads. We’re also introducing Classic Plus, a new face model, along with Extracted Key Frames for more accurate detection in videos. And for Pro users, you can now keep the original video specs, so your final output stays production-ready. Less manual work. Better results. More creative freedom. Try the new Face Swap on Akool and see the difference. 🚀
0:42

Face swapping just leveled up. Akool Face Swap just got a major upgrade — designed to make AI video creation faster, smarter, and way more powerful. Now you can process up to 50 files in batch, swap faces directly on GIFs, and benefit from smarter face detection that automatically removes duplicate faces across uploads. We’re also introducing Classic Plus, a new face model, along with Extracted Key Frames for more accurate detection in videos. And for Pro users, you can now keep the original video specs, so your final output stays production-ready. Less manual work. Better results. More creative freedom. Try the new Face Swap on Akool and see the difference. 🚀

Akool Inc

298,768 Aufrufe • vor 5 Monaten

WorldFlow3D: Unbounded 3D World Generation 🌍 by Flow Through Hierarchical Distributions, without VAEs ! We reformulate 3D generation as flowing through sequentially finer 3D distributions, cutting training time by more than half ⏱️ compared to existing approaches! Vectorized map layouts provide full scene controllability 🗺️, and a novel flow-field alignment process enables causally coherent, spatially unbounded generation 🌍. This generative method generalizes across both real and synthetic data distributions! Project: Project led by Amogh Joshi and Julian Ost — will be super fun to build on this! 🔥
0:53

WorldFlow3D: Unbounded 3D World Generation 🌍 by Flow Through Hierarchical Distributions, without VAEs ! We reformulate 3D generation as flowing through sequentially finer 3D distributions, cutting training time by more than half ⏱️ compared to existing approaches! Vectorized map layouts provide full scene controllability 🗺️, and a novel flow-field alignment process enables causally coherent, spatially unbounded generation 🌍. This generative method generalizes across both real and synthetic data distributions! Project: Project led by Amogh Joshi and Julian Ost — will be super fun to build on this! 🔥

Felix Heide

19,534 Aufrufe • vor 4 Monaten

How well do egocentric human videos scale? Apparently, really well. Dyna Robotics has introduced Dyna-2, a world-action model (WAM) pre-trained on more than 1 million hours of real egocentric human video. It learns by jointly predicting the next video frames and the actions, and that video prediction (world modeling) is what carries learning from humans to robots. • More human video predictably improves action prediction. • Task accuracy improves across embodiments (bimanual arms and a humanoid) with zero robot data in pre-training. • Post-trained on just a few hours of robot data, mean task performance rose 20% to 53% as human pre-training scaled 1k to 1M hours. WAM vs VLA, matched apple-to-apple: the WAM hit 1.55x the VLA's success rate and won 65% of head-to-heads. Deployed zero-shot at unseen customer sites, it passed 87% vs the VLA's 46%.
0:28

How well do egocentric human videos scale? Apparently, really well. Dyna Robotics has introduced Dyna-2, a world-action model (WAM) pre-trained on more than 1 million hours of real egocentric human video. It learns by jointly predicting the next video frames and the actions, and that video prediction (world modeling) is what carries learning from humans to robots. • More human video predictably improves action prediction. • Task accuracy improves across embodiments (bimanual arms and a humanoid) with zero robot data in pre-training. • Post-trained on just a few hours of robot data, mean task performance rose 20% to 53% as human pre-training scaled 1k to 1M hours. WAM vs VLA, matched apple-to-apple: the WAM hit 1.55x the VLA's success rate and won 65% of head-to-heads. Deployed zero-shot at unseen customer sites, it passed 87% vs the VLA's 46%.

The Humanoid Hub

15,034 Aufrufe • vor 17 Tagen

Ethiopia is actively expanding and upgrading its road network to improve connectivity, boost trade, and support economic growth. Current initiatives focus on constructing new highways, rural roads, and urban streets, as well as modernising existing routes to reduce travel times, improve safety, and lower transportation costs. These efforts enhance access to schools, health services, and markets, while supporting industrialisation, agriculture, and local businesses. By prioritising both construction and maintenance, Ethiopia is creating a more efficient, reliable, and sustainable road system that strengthens regional integration and improves the daily lives of communities across the country. #PMOEthiopia #VisionOfProsperity #የኢትዮጵያብልጽግናራእይ
0:30

Ethiopia is actively expanding and upgrading its road network to improve connectivity, boost trade, and support economic growth. Current initiatives focus on constructing new highways, rural roads, and urban streets, as well as modernising existing routes to reduce travel times, improve safety, and lower transportation costs. These efforts enhance access to schools, health services, and markets, while supporting industrialisation, agriculture, and local businesses. By prioritising both construction and maintenance, Ethiopia is creating a more efficient, reliable, and sustainable road system that strengthens regional integration and improves the daily lives of communities across the country. #PMOEthiopia #VisionOfProsperity #የኢትዮጵያብልጽግናራእይ

Office of the Prime Minister - Ethiopia

16,072 Aufrufe • vor 7 Monaten

4DGT: Learning a 4D Gaussian Transformer Using Real-World Monocular Videos Abstract: We propose 4DGT, a 4D Gaussian-based Transformer model for dynamic scene reconstruction, trained entirely on real-world monocular posed videos. Using 4D Gaussian as an inductive bias, 4DGT unifies static and dynamic components, enabling the modeling of complex, time-varying environments with varying object lifespans. We introduced a novel density control strategy in training, which allows our 4DGT to handle longer space-time input while maintaining efficient rendering at runtime. Our model processes 64 consecutive posed frames in a rolling-window fashion, predicting consistent 4D Gaussians in the scene. Unlike optimization-based methods, 4DGT performs purely feed-forward inference, reducing reconstruction time from hours to seconds and scaling effectively to long video sequences. Trained only on large-scale monocular posed video datasets, 4DGT can significantly outperform prior Gaussian-based networks in real-world videos and achieve on-par accuracy with optimization-based methods on cross-domain videos.
5:19

4DGT: Learning a 4D Gaussian Transformer Using Real-World Monocular Videos Abstract: We propose 4DGT, a 4D Gaussian-based Transformer model for dynamic scene reconstruction, trained entirely on real-world monocular posed videos. Using 4D Gaussian as an inductive bias, 4DGT unifies static and dynamic components, enabling the modeling of complex, time-varying environments with varying object lifespans. We introduced a novel density control strategy in training, which allows our 4DGT to handle longer space-time input while maintaining efficient rendering at runtime. Our model processes 64 consecutive posed frames in a rolling-window fashion, predicting consistent 4D Gaussians in the scene. Unlike optimization-based methods, 4DGT performs purely feed-forward inference, reducing reconstruction time from hours to seconds and scaling effectively to long video sequences. Trained only on large-scale monocular posed video datasets, 4DGT can significantly outperform prior Gaussian-based networks in real-world videos and achieve on-par accuracy with optimization-based methods on cross-domain videos.

MrNeRF

34,782 Aufrufe • vor 1 Jahr

It’s INSANE how easy it is to Create Photorealistic Characters and Animate them within Minutes using Unreal Engine. Ideation to Render now takes a few hours ONLY and that’s super liberating for artists who want to tell stories 🤯 With Unreal you can now create facial animations for your Metahuman just by using audio without any complex setups, or you can even use your Iphone and Webcam to drive the face animations. It uses AI to predict facial muscle movements, and the results are honestly insane. This is a total game-changer, especially for filmmakers, storytellers and even game developers.
2:05

It’s INSANE how easy it is to Create Photorealistic Characters and Animate them within Minutes using Unreal Engine. Ideation to Render now takes a few hours ONLY and that’s super liberating for artists who want to tell stories 🤯 With Unreal you can now create facial animations for your Metahuman just by using audio without any complex setups, or you can even use your Iphone and Webcam to drive the face animations. It uses AI to predict facial muscle movements, and the results are honestly insane. This is a total game-changer, especially for filmmakers, storytellers and even game developers.

Bad Decisions Studio

15,201 Aufrufe • vor 1 Jahr