Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

this looks insane, snapchat just dropped Dynamic Concepts Personalization from Single Videos propose a new technique for personalizing text-to-video models, enabling them to capture, manipulate and combine Dynamic Concepts.

AK

512,052 subscribers

29,236 views • 1 year ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

5 Comments

AK1 year ago

discuss:

The Information1 year ago

A recent production hiccup with Nvidia’s latest AI chip shows the growing pressures in its relationship with its main chipmaking partner, TSMC.

0x_Vivek1 year ago

dynamic concepts are cool, but personalization at scale hinges on model efficiency and data diversity. this is just the start.

Universa1 year ago

@_akhaliq Exciting to see Snapchat's introduction of Dynamic Concepts Personalization from Single Videos. This innovation is exactly what the Universa Hackathon, running from January to February 2025, is all about.

DisruptiveConcepts1 year ago

Have a great day

Related Videos

Designing an Encoder for Fast Personalization of Text-to-Image Models TL;DR: use an encoder to personalize a text-to-image model to new concepts with a single image and 5-15 tuning steps abs: project page:

Designing an Encoder for Fast Personalization of Text-to-Image Models TL;DR: use an encoder to personalize a text-to-image model to new concepts with a single image and 5-15 tuning steps abs: project page:

AK

165,168 views • 3 years ago

[SIGGRAPH ASIA '25] Detail-Enhanced Gaussian Splatting for Large-Scale Volumetric Capture Contributions: - A two-stage approach to performance capture, combining a scene-scale capture rig and a single-actor facial capture rig. - A novel high-quality scene-scale volumetric performance capture rig, incorporating both static and dynamic cameras to track the performance of multiple actors. - A reconstruction pipeline for dynamic performance capture, featuring stable calibration of moving cameras and 4DGS with improved dynamic range and color fidelity. - A detail enhancement Diffusion Model, which supports 4K, RGB, and Alpha, with improved temporal stability.

[SIGGRAPH ASIA '25] Detail-Enhanced Gaussian Splatting for Large-Scale Volumetric Capture Contributions: - A two-stage approach to performance capture, combining a scene-scale capture rig and a single-actor facial capture rig. - A novel high-quality scene-scale volumetric performance capture rig, incorporating both static and dynamic cameras to track the performance of multiple actors. - A reconstruction pipeline for dynamic performance capture, featuring stable calibration of moving cameras and 4DGS with improved dynamic range and color fidelity. - A detail enhancement Diffusion Model, which supports 4K, RGB, and Alpha, with improved temporal stability.

MrNeRF

42,456 views • 9 months ago

VSTAR Generative Temporal Nursing for Longer Dynamic Video Synthesis Despite tremendous progress in the field of text-to-video (T2V) synthesis, open-sourced T2V diffusion models struggle to generate longer videos with dynamically varying and evolving content. They tend to

VSTAR Generative Temporal Nursing for Longer Dynamic Video Synthesis Despite tremendous progress in the field of text-to-video (T2V) synthesis, open-sourced T2V diffusion models struggle to generate longer videos with dynamically varying and evolving content. They tend to

AK

36,982 views • 2 years ago

OK, this is insane.. Alibaba just dropped 4 new AI models at Apsara 2025, and they’re wild: → a 1 trillion parameter LLM → a vision model that codes from images → an omni-model for text/audio/video → and a new Wan 2.5 preview for video + audio gen more details below:👇

OK, this is insane.. Alibaba just dropped 4 new AI models at Apsara 2025, and they’re wild: → a 1 trillion parameter LLM → a vision model that codes from images → an omni-model for text/audio/video → and a new Wan 2.5 preview for video + audio gen more details below:👇

Hamza Khalid

26,707 views • 10 months ago

New short course: Building Multimodal Search and RAG", by Weaviate AI Database's Sebastia(N_) Witalec ✊🏽✊🏾✊🏿. Contrastive learning is used to train models to map vectors into an embedding space by pulling similar concepts closer together and pushing dissimilar concepts away from each other. This technique is also used to train multimodal embedding models that capture semantic similarity across different modalities like text, images, and audio. These multimodal embeddings can be used to build multimodal search and RAG systems. In this course, you'll learn how contrastive learning works, and how to add multimodality to RAG – so your models can draw on diverse, relevant context to answer questions. For example, a query about a financial report might synthesize information from text snippets, graphs, tables, and slides. You will also learn how visual instruction tuning lets you integrate image understanding into language models, and build a multi-vector recommender system using Weaviate’s open-source vector database. Please sign up here:

New short course: Building Multimodal Search and RAG", by Weaviate AI Database's Sebastia(N_) Witalec ✊🏽✊🏾✊🏿. Contrastive learning is used to train models to map vectors into an embedding space by pulling similar concepts closer together and pushing dissimilar concepts away from each other. This technique is also used to train multimodal embedding models that capture semantic similarity across different modalities like text, images, and audio. These multimodal embeddings can be used to build multimodal search and RAG systems. In this course, you'll learn how contrastive learning works, and how to add multimodality to RAG – so your models can draw on diverse, relevant context to answer questions. For example, a query about a financial report might synthesize information from text snippets, graphs, tables, and slides. You will also learn how visual instruction tuning lets you integrate image understanding into language models, and build a multi-vector recommender system using Weaviate’s open-source vector database. Please sign up here:

Andrew Ng

104,371 views • 2 years ago

Beyond Scene Ingredients, Pika 2.0 also includes upgrades to our text-to-video and image-to-video models, delivering higher visual quality, more dynamic motion, and better text alignment. Try Pika 2.0 for free until December 22 at pika dot art. Credit: Ilor Elmaleh

Beyond Scene Ingredients, Pika 2.0 also includes upgrades to our text-to-video and image-to-video models, delivering higher visual quality, more dynamic motion, and better text alignment. Try Pika 2.0 for free until December 22 at pika dot art. Credit: Ilor Elmaleh

Pika

41,928 views • 1 year ago

Google announces PALP Prompt Aligned Personalization of Text-to-Image Models paper page: Content creators often aim to create personalized images using personal subjects that go beyond the capabilities of conventional text-to-image models. Additionally, they may want the resulting image to encompass a specific location, style, ambiance, and more. Existing personalization methods may compromise personalization ability or the alignment to complex textual prompts. This trade-off can impede the fulfillment of user prompts and subject fidelity. We propose a new approach focusing on personalization methods for a single prompt to address this issue. We term our approach prompt-aligned personalization. While this may seem restrictive, our method excels in improving text alignment, enabling the creation of images with complex and intricate prompts, which may pose a challenge for current techniques. In particular, our method keeps the personalized model aligned with a target prompt using an additional score distillation sampling term. We demonstrate the versatility of our method in multi- and single-shot settings and further show that it can compose multiple subjects or use inspiration from reference images, such as artworks. We compare our approach quantitatively and qualitatively with existing baselines and state-of-the-art techniques.

Google announces PALP Prompt Aligned Personalization of Text-to-Image Models paper page: Content creators often aim to create personalized images using personal subjects that go beyond the capabilities of conventional text-to-image models. Additionally, they may want the resulting image to encompass a specific location, style, ambiance, and more. Existing personalization methods may compromise personalization ability or the alignment to complex textual prompts. This trade-off can impede the fulfillment of user prompts and subject fidelity. We propose a new approach focusing on personalization methods for a single prompt to address this issue. We term our approach prompt-aligned personalization. While this may seem restrictive, our method excels in improving text alignment, enabling the creation of images with complex and intricate prompts, which may pose a challenge for current techniques. In particular, our method keeps the personalized model aligned with a target prompt using an additional score distillation sampling term. We demonstrate the versatility of our method in multi- and single-shot settings and further show that it can compose multiple subjects or use inspiration from reference images, such as artworks. We compare our approach quantitatively and qualitatively with existing baselines and state-of-the-art techniques.

AK

90,199 views • 2 years ago

Midjourney v7 plus Luma new Camera Motion Concepts is incredible. This is Mercs. I created this short last night, using my personalization settings in the new MJ v7 along with the new dynamic camera options in #Ray2 The camera moves you can get are next level. I used the new Actor Mode for voice and Dreamina AI for impressive lip sync. It's been incredible building a film studio with Promise. It forces me to stay on my AI filmmaker grind and try out the latest and greatest toolstack. Images upscaled in Magnific.ai and videos in Topaz Labs Edited in Adobe Premiere.

Midjourney v7 plus Luma new Camera Motion Concepts is incredible. This is Mercs. I created this short last night, using my personalization settings in the new MJ v7 along with the new dynamic camera options in #Ray2 The camera moves you can get are next level. I used the new Actor Mode for voice and Dreamina AI for impressive lip sync. It's been incredible building a film studio with Promise. It forces me to stay on my AI filmmaker grind and try out the latest and greatest toolstack. Images upscaled in Magnific.ai and videos in Topaz Labs Edited in Adobe Premiere.

Dave Clark

31,744 views • 1 year ago

📢WorldAgents: 3D worlds only from 2D image models - without any training! We propose an agentic approach with a Director (VLM) to plan the scene, a Generator (Flux or NanoBanana) for new views, and a Verifier (VLM) for selection / 3D consistency. -> High-fidelity 3D worlds from a single text prompt. What's remarkable: our agents find consistent views from 2D image models to obtain 3D-consistent worlds; this shows that image models contain world priors - agents just need to find them! Great work by Ziya Erkoç Angela Dai

📢WorldAgents: 3D worlds only from 2D image models - without any training! We propose an agentic approach with a Director (VLM) to plan the scene, a Generator (Flux or NanoBanana) for new views, and a Verifier (VLM) for selection / 3D consistency. -> High-fidelity 3D worlds from a single text prompt. What's remarkable: our agents find consistent views from 2D image models to obtain 3D-consistent worlds; this shows that image models contain world priors - agents just need to find them! Great work by Ziya Erkoç Angela Dai

Matthias Niessner

19,005 views • 4 months ago

This is Text-to-CAD. A new open-source prompt interface for generating CAD files through text prompts. It looks insane 🤯

This is Text-to-CAD. A new open-source prompt interface for generating CAD files through text prompts. It looks insane 🤯

Alvaro Cintas

111,396 views • 1 year ago

5 Out Offense Video: NBA Concepts From Championship Teams REPLY below for a free PDF with the 5 Out Offensive Concepts. In this video, you'll learn how to invigorate your offense by incorporating these 5 Out Offense concepts. These simple to advanced concepts were used by recent NBA champions, including the Denver Nuggets and the Golden State Warriors. Former Guam National Team Coach, Brent Tipton, compiled these clips. He has used them with teams ranging from youth to professional levels. No matter the age you're coaching, these concepts can help your players generate efficient offense! By implementing these offensive concepts, your team will create easy scoring opportunities as you flow from transition into the half court. No matter how the defense guards you, these 5-out offense concepts provide a solution that results in quality shots.

5 Out Offense Video: NBA Concepts From Championship Teams REPLY below for a free PDF with the 5 Out Offensive Concepts. In this video, you'll learn how to invigorate your offense by incorporating these 5 Out Offense concepts. These simple to advanced concepts were used by recent NBA champions, including the Denver Nuggets and the Golden State Warriors. Former Guam National Team Coach, Brent Tipton, compiled these clips. He has used them with teams ranging from youth to professional levels. No matter the age you're coaching, these concepts can help your players generate efficient offense! By implementing these offensive concepts, your team will create easy scoring opportunities as you flow from transition into the half court. No matter how the defense guards you, these 5-out offense concepts provide a solution that results in quality shots.

Joe Haefner | Breakthrough Basketball

45,242 views • 1 year ago

These AI celebrity videos are a weirdly effective way to learn new concepts. And insanely viral - this one has 5M views in two days 🤯

These AI celebrity videos are a weirdly effective way to learn new concepts. And insanely viral - this one has 5M views in two days 🤯

Justine Moore

2,443,649 views • 1 year ago

Turn any celebrity selfie into a dynamic AI video instantly! This PixelFlow combines Nano Banana Pro and Kling 2.5 Turbo to create hyper-realistic, expressive celebrity-style selfie videos from a single image expressive, Try it now →

Turn any celebrity selfie into a dynamic AI video instantly! This PixelFlow combines Nano Banana Pro and Kling 2.5 Turbo to create hyper-realistic, expressive celebrity-style selfie videos from a single image expressive, Try it now →

Segmind

20,086 views • 7 months ago

Make Pixels Dance: High-Dynamic Video Generation paper page: Creating high-dynamic videos such as motion-rich actions and sophisticated visual effects poses a significant challenge in the field of artificial intelligence. Unfortunately, current state-of-the-art video generation methods, primarily focusing on text-to-video generation, tend to produce video clips with minimal motions despite maintaining high fidelity. We argue that relying solely on text instructions is insufficient and suboptimal for video generation. In this paper, we introduce PixelDance, a novel approach based on diffusion models that incorporates image instructions for both the first and last frames in conjunction with text instructions for video generation. Comprehensive experimental results demonstrate that PixelDance trained with public data exhibits significantly better proficiency in synthesizing videos with complex scenes and intricate motions, setting a new standard for video generation.

Make Pixels Dance: High-Dynamic Video Generation paper page: Creating high-dynamic videos such as motion-rich actions and sophisticated visual effects poses a significant challenge in the field of artificial intelligence. Unfortunately, current state-of-the-art video generation methods, primarily focusing on text-to-video generation, tend to produce video clips with minimal motions despite maintaining high fidelity. We argue that relying solely on text instructions is insufficient and suboptimal for video generation. In this paper, we introduce PixelDance, a novel approach based on diffusion models that incorporates image instructions for both the first and last frames in conjunction with text instructions for video generation. Comprehensive experimental results demonstrate that PixelDance trained with public data exhibits significantly better proficiency in synthesizing videos with complex scenes and intricate motions, setting a new standard for video generation.

AK

102,444 views • 2 years ago

Imagine batch generating 50 packaging concepts for a new product all in one session. With Manus, you can: 👉 Generate dozens of design directions (styles, labels, aesthetics) 👉 Format them directly into a polished deck 👉 Ready-to-use images downloaded into a single file No bottlenecks. No scattered files, just moving from rapid iteration to clear direction. See what packaging exploration looks when it’s this fast!

Imagine batch generating 50 packaging concepts for a new product all in one session. With Manus, you can: 👉 Generate dozens of design directions (styles, labels, aesthetics) 👉 Format them directly into a polished deck 👉 Ready-to-use images downloaded into a single file No bottlenecks. No scattered files, just moving from rapid iteration to clear direction. See what packaging exploration looks when it’s this fast!

ManusAI

13,952 views • 11 months ago

ByteDance just unveiled HuMo, a unified framework for human-centric video generation. With HuMo you can: • Generate videos from text + image • Create audio-synced videos from text + audio • Combine text, image, and audio for maximum control • Preserve consistent subjects across frames • Achieve natural audio-visual synchronization This isn’t just another release , it’s a big step toward fully controllable, coherent, and cinematic AI video.

ByteDance just unveiled HuMo, a unified framework for human-centric video generation. With HuMo you can: • Generate videos from text + image • Create audio-synced videos from text + audio • Combine text, image, and audio for maximum control • Preserve consistent subjects across frames • Achieve natural audio-visual synchronization This isn’t just another release , it’s a big step toward fully controllable, coherent, and cinematic AI video.

DStudioproject

29,408 views • 11 months ago

We just dropped the tutorials showing how incredibly easy it is to install and customize BeatScreen! A dynamic screen for #VRChat that reacts in real time to video and music using AudioLink. Check it out 👇

We just dropped the tutorials showing how incredibly easy it is to install and customize BeatScreen! A dynamic screen for #VRChat that reacts in real time to video and music using AudioLink. Check it out 👇

DX

34,924 views • 4 months ago

v5 is insane for comment-farming setups. The goal isn’t “perfect” ads — it’s good enough to slip into the feed and introduce a problem/solution concept. With this workflow the concepts are basically endless, and engagement goes crazy. One of my videos just hit 19K shares.

v5 is insane for comment-farming setups. The goal isn’t “perfect” ads — it’s good enough to slip into the feed and introduce a problem/solution concept. With this workflow the concepts are basically endless, and engagement goes crazy. One of my videos just hit 19K shares.

Franky Shaw

18,518 views • 5 months ago

here's how to do this in 2 minutes with ai: - generate character with nb2 - clone voice with elevenlabs - lipsync with infinitetalk - that's it congrats you just saved $500 and 2 weeks of "creative concepts" from someone who points at text for a living.

here's how to do this in 2 minutes with ai: - generate character with nb2 - clone voice with elevenlabs - lipsync with infinitetalk - that's it congrats you just saved $500 and 2 weeks of "creative concepts" from someone who points at text for a living.

0x ROAS

51,309 views • 7 months ago