Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Ditto Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset

_akhaliq's profile picture

AK

519,666 subscribers

18,082 görüntüleme • 11 ay önce •via X (Twitter)

EğitimHaberler & PolitikaBilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

GaussianSpeech: Audio-Driven Gaussian Avatars Contributions: • The first transformer-based sequence model for audio-driven head animation synthesis of a lightweight 3DGS based avatar. By animating our optimized 3DGS avatar directly with our transformer model, we achieve temporally coherent animation sequences while characterizing fine-scale face details and speaker-specific style. • A new high-quality audio-video dataset, comprising high-resolution 16-view dataset of 6 native English speakers (Standard American & British). The dataset has a total of 2500 sequences, with overall recordings of ∼3.5 hours.
1:06

GaussianSpeech: Audio-Driven Gaussian Avatars Contributions: • The first transformer-based sequence model for audio-driven head animation synthesis of a lightweight 3DGS based avatar. By animating our optimized 3DGS avatar directly with our transformer model, we achieve temporally coherent animation sequences while characterizing fine-scale face details and speaker-specific style. • A new high-quality audio-video dataset, comprising high-resolution 16-view dataset of 6 native English speakers (Standard American & British). The dataset has a total of 2500 sequences, with overall recordings of ∼3.5 hours.

MrNeRF

11,476 görüntüleme • 1 yıl önce

high quality video for a high quality song
0:59

high quality video for a high quality song

໊

1,041,233 görüntüleme • 8 ay önce

high quality video for a high quality song
0:52

high quality video for a high quality song

ْ

130,641 görüntüleme • 4 ay önce

high quality video for a high quality song
2:01

high quality video for a high quality song

ْ

84,993 görüntüleme • 9 ay önce

🚀 Unitree open-sources UnifoLM-WBT-Dataset — a high-quality real-world humanoid robot whole-body teleoperation (WBT) dataset for open environments. 🥳Publicly available since March 5, 2026, the dataset will continue to receive high-frequency rolling updates. It aims to establish the most comprehensive real-world humanoid robot dataset in terms of scenario coverage, task complexity, and manipulation diversity. 👉 Explore the dataset here:
1:27

🚀 Unitree open-sources UnifoLM-WBT-Dataset — a high-quality real-world humanoid robot whole-body teleoperation (WBT) dataset for open environments. 🥳Publicly available since March 5, 2026, the dataset will continue to receive high-frequency rolling updates. It aims to establish the most comprehensive real-world humanoid robot dataset in terms of scenario coverage, task complexity, and manipulation diversity. 👉 Explore the dataset here:

Unitree

5,633,606 görüntüleme • 6 ay önce

Today we’re sharing two new advances in our generative AI research: Emu Video & Emu Edit. Details ➡️ These new models deliver exciting results in high quality, diffusion-based text-to-video generation & controlled image editing w/ text instructions. 🧵
0:33

Today we’re sharing two new advances in our generative AI research: Emu Video & Emu Edit. Details ➡️ These new models deliver exciting results in high quality, diffusion-based text-to-video generation & controlled image editing w/ text instructions. 🧵

AI at Meta

798,500 görüntüleme • 2 yıl önce

Meta FAIR recently released the Seamless Interaction Dataset, the largest known high-quality video dataset of its kind, with: 4,000+ diverse participants 4,000+ hours of footage 65k+ interactions 5,000+ annotated samples This dataset of full-body, in-person, face-to-face interaction videos represents a crucial stepping stone to understanding and modeling how people communicate and behave when they’re together—advancing AI's ability to generate more natural conversations and human-like gestures. Download the dataset on Hugging Face: Learn more about the dataset:
1:05

Meta FAIR recently released the Seamless Interaction Dataset, the largest known high-quality video dataset of its kind, with: 4,000+ diverse participants 4,000+ hours of footage 65k+ interactions 5,000+ annotated samples This dataset of full-body, in-person, face-to-face interaction videos represents a crucial stepping stone to understanding and modeling how people communicate and behave when they’re together—advancing AI's ability to generate more natural conversations and human-like gestures. Download the dataset on Hugging Face: Learn more about the dataset:

AI at Meta

23,836 görüntüleme • 1 yıl önce

🚀 Introducing DiffusionRenderer, a neural rendering engine powered by video diffusion models. 🎥 Estimates high-quality geometry and materials from videos, synthesizes photorealistic light transport, enables relighting and material editing with realistic shadows and reflections
2:14

🚀 Introducing DiffusionRenderer, a neural rendering engine powered by video diffusion models. 🎥 Estimates high-quality geometry and materials from videos, synthesizes photorealistic light transport, enables relighting and material editing with realistic shadows and reflections

Zian Wang

125,619 görüntüleme • 1 yıl önce

pip install dria All you need to generate high-quality, grounded synthetic data. Experiment with different models, incorporate tool usage, and generate at scale in just a few lines of code.
0:52

pip install dria All you need to generate high-quality, grounded synthetic data. Experiment with different models, incorporate tool usage, and generate at scale in just a few lines of code.

Dria

19,351 görüntüleme • 1 yıl önce

🚨🚨 New #SLAM #RealityCapture Dataset! 🚨🚨 The Oxford Spires Dataset. New mapping benchmark in Oxford's famous colleges and Blenheim Palace focusing on high quality reconstruction (#nerf #gaussiansplatting) and colorization. #spatialai #oxford
1:06

🚨🚨 New #SLAM #RealityCapture Dataset! 🚨🚨 The Oxford Spires Dataset. New mapping benchmark in Oxford's famous colleges and Blenheim Palace focusing on high quality reconstruction (#nerf #gaussiansplatting) and colorization. #spatialai #oxford

Oxford Dynamic Robot Systems

18,291 görüntüleme • 1 yıl önce

Meta announces Movie Gen A Cast of Media Foundation Models We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user’s image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models
0:30

Meta announces Movie Gen A Cast of Media Foundation Models We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user’s image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models

AK

62,719 görüntüleme • 2 yıl önce

It’s amazing to see how engaged our students are with the high quality instruction our teachers provide. Excellent work, Condit Elementary and Bastian ES #HISD #LeadOn #Schoolvisits
0:56

It’s amazing to see how engaged our students are with the high quality instruction our teachers provide. Excellent work, Condit Elementary and Bastian ES #HISD #LeadOn #Schoolvisits

Mike Miles

16,988 görüntüleme • 3 yıl önce

FasterCache Training-Free Video Diffusion Model Acceleration with High Quality
1:10

FasterCache Training-Free Video Diffusion Model Acceleration with High Quality

AK

91,309 görüntüleme • 1 yıl önce

MatAnyone 2 is out on Hugging Face Scaling Video Matting via a Learned Quality Evaluator paper: app:
0:40

MatAnyone 2 is out on Hugging Face Scaling Video Matting via a Learned Quality Evaluator paper: app:

AK

34,927 görüntüleme • 7 ay önce

Shape your video scene, start to finish. Choose your first and last frames, or create seamless loops. With improved instruction following and higher video quality.
0:33

Shape your video scene, start to finish. Choose your first and last frames, or create seamless loops. With improved instruction following and higher video quality.

Grok Imagine

222,223 görüntüleme • 1 ay önce

Reference Hub. Creativity Unleashed. Now live. The world’s first AI video reference-based creation community 🎬 A new creative paradigm: powered by Vidu’s reference-based video generation, creators can select from different reference types to produce professional-quality videos 🌏 8 reference types: professional-level video results without complex prompting 🎞 Making professional output effortless: stable, high-quality results with greater efficiency #YesVidu #Viduai #Creativity #ReferenceHub
1:31

Reference Hub. Creativity Unleashed. Now live. The world’s first AI video reference-based creation community 🎬 A new creative paradigm: powered by Vidu’s reference-based video generation, creators can select from different reference types to produce professional-quality videos 🌏 8 reference types: professional-level video results without complex prompting 🎞 Making professional output effortless: stable, high-quality results with greater efficiency #YesVidu #Viduai #Creativity #ReferenceHub

Vidu AI

14,942 görüntüleme • 8 ay önce

🚨 Wan 2.7 is now live on fal! ⚡ Big upgrades in visuals, motion, audio, style, and consistency 🎬 First & last frame, 9-grid I2V, subject + voice reference 🎨 Instruction-based editing and video recreation
0:20

🚨 Wan 2.7 is now live on fal! ⚡ Big upgrades in visuals, motion, audio, style, and consistency 🎬 First & last frame, 9-grid I2V, subject + voice reference 🎨 Instruction-based editing and video recreation

fal

27,414 görüntüleme • 6 ay önce

Pushed a small but mighty update to Claude Engineer. When using editing, Claude will select the batches based only on the code blocks worth editing. Showing the selected blocks and the proposed update. This should improve editing quality by a lot and token management too.
1:08

Pushed a small but mighty update to Claude Engineer. When using editing, Claude will select the batches based only on the code blocks worth editing. Showing the selected blocks and the proposed update. This should improve editing quality by a lot and token management too.

Pietro Schirano

48,324 görüntüleme • 2 yıl önce

Low quality video High quality performance ❤️🇮🇳🫡
1:41

Low quality video High quality performance ❤️🇮🇳🫡

मुसाफिर🇮🇳

12,055 görüntüleme • 1 ay önce