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。

Introducing ConceptAttention, an approach to interpreting diffusion transformer models! Write a prompt, choose some concepts, generate an image, and get high-quality heatmaps of text concepts. Our method outperforms existing methods like cross attention. Link to demo 👇

Alec Helbling

10,728 subscribers

36,631 次观看 • 1 年前 •via X (Twitter)

科学技术新闻政治教育

Anya Rossi• Live Now

Private livecam show

11 条评论

Alec Helbling 的头像

Alec Helbling1 年前

We have a live interactive demo hosted on Huggingface Spaces:

Alec Helbling 的头像

Alec Helbling1 年前

Check out the code here:

Alec Helbling 的头像

Alec Helbling1 年前

We repurpose the parameters of multi-modal DiT models (i.e. Flux) without training to create rich contextualized embeddings of text concepts. This allows us to create high quality saliency maps. We wrote a paper about our method:

Rainmaker 的头像

Rainmaker2 年前

Join me as I put several Machine Learning models head-to-head to see which one can beat the market and deliver strong returns. In this free Substack post I share several models that deliver better returns with much lower drawdown compared to Buy-and-Hold approach.

Rishi 的头像

Rishi1 年前

Very Nice Idea, Explainable AI as a field is not that much explored so nice to see good work in that domain

Rishi 的头像

Rishi1 年前

Will this work for not non Diffusion based models ?

Minh Nhat Nguyen 的头像

Minh Nhat Nguyen1 年前

i am ... going to see how well this works for video

 007 的头像

 0071 年前

Cool

Julien Blanchon 🇺🇦 的头像

Julien Blanchon 🇺🇦1 年前

Curious about what your intuition about the entanglement between dog and cat features ?

Alec Helbling 的头像

Alec Helbling1 年前

Absolutely. Our observations have been that this approach works very well for discerning distinct features (like dog and background) but struggles with examples like you show where you have two very similar concepts. There is clearly some more complex mechanism that allows the model to differentiate between these concepts that unfortunately our approach alone is not able to discern. It is worth noting this limitation is also at play in cross attention mechanisms, and poor object attribute assignment is a known limitation of current diffusion models.

Latent Spacer 的头像

Latent Spacer1 年前

I learned a lot from the paper, great work 👏

相关视频

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

165,151 次观看 • 3 年前

3DTopia-XL High-Quality 3D PBR Asset Generation via Primitive Diffusion demo: model: 3DTopia-XL scales high-quality 3D asset generation using Diffusion Transformer (DiT) built upon an expressive and efficient 3D representation, PrimX. The denoising process takes 5 seconds to generate a 3D PBR asset from text/image input which is ready for the graphics pipeline to use.

3DTopia-XL High-Quality 3D PBR Asset Generation via Primitive Diffusion demo: model: 3DTopia-XL scales high-quality 3D asset generation using Diffusion Transformer (DiT) built upon an expressive and efficient 3D representation, PrimX. The denoising process takes 5 seconds to generate a 3D PBR asset from text/image input which is ready for the graphics pipeline to use.

AK

87,086 次观看 • 1 年前

Presto! Distilling Steps and Layers for Accelerating Music Generation Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers via reducing both sampling steps and cost per step. To reduce steps, we develop a new score-based distribution matching distillation (DMD) method for the EDM-family of diffusion models, the first GAN-based distillation method for TTM. To reduce the cost per step, we develop a simple, but powerful improvement to a recent layer distillation method that improves learning via better preserving hidden state variance. Finally, we combine our step and layer distillation methods together for a dual-faceted approach. We evaluate our step and layer distillation methods independently and show each yield best-in-class performance. Our combined distillation method can generate high-quality outputs with improved diversity, accelerating our base model by 10-18x (230/435ms latency for 32 second mono/stereo 44.1kHz, 15x faster than comparable SOTA) -- the fastest high-quality TTM to our knowledge.

Presto! Distilling Steps and Layers for Accelerating Music Generation Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers via reducing both sampling steps and cost per step. To reduce steps, we develop a new score-based distribution matching distillation (DMD) method for the EDM-family of diffusion models, the first GAN-based distillation method for TTM. To reduce the cost per step, we develop a simple, but powerful improvement to a recent layer distillation method that improves learning via better preserving hidden state variance. Finally, we combine our step and layer distillation methods together for a dual-faceted approach. We evaluate our step and layer distillation methods independently and show each yield best-in-class performance. Our combined distillation method can generate high-quality outputs with improved diversity, accelerating our base model by 10-18x (230/435ms latency for 32 second mono/stereo 44.1kHz, 15x faster than comparable SOTA) -- the fastest high-quality TTM to our knowledge.

AK

30,430 次观看 • 1 年前

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

90,188 次观看 • 2 年前

Our controlNET scribble preprocessor takes a flat sketch and sends it to our skybox diffusion to return an image based on a prompt that follows shapes and concepts. Still in testing, but in short, a user draws sketch, sends request, gets skybox that fits sketch. #texttoskybox

Our controlNET scribble preprocessor takes a flat sketch and sends it to our skybox diffusion to return an image based on a prompt that follows shapes and concepts. Still in testing, but in short, a user draws sketch, sends request, gets skybox that fits sketch. #texttoskybox

Blockade Labs

96,893 次观看 • 3 年前

TurboEdit Instant text-based image editing discuss: We address the challenges of precise image inversion and disentangled image editing in the context of few-step diffusion models. We introduce an encoder based iterative inversion technique. The inversion network is conditioned on the input image and the reconstructed image from the previous step, allowing for correction of the next reconstruction towards the input image. We demonstrate that disentangled controls can be easily achieved in the few-step diffusion model by conditioning on an (automatically generated) detailed text prompt. To manipulate the inverted image, we freeze the noise maps and modify one attribute in the text prompt (either manually or via instruction based editing driven by an LLM), resulting in the generation of a new image similar to the input image with only one attribute changed. It can further control the editing strength and accept instructive text prompt. Our approach facilitates realistic text-guided image edits in real-time, requiring only 8 number of functional evaluations (NFEs) in inversion (one-time cost) and 4 NFEs per edit. Our method is not only fast, but also significantly outperforms state-of-the-art multi-step diffusion editing techniques.

TurboEdit Instant text-based image editing discuss: We address the challenges of precise image inversion and disentangled image editing in the context of few-step diffusion models. We introduce an encoder based iterative inversion technique. The inversion network is conditioned on the input image and the reconstructed image from the previous step, allowing for correction of the next reconstruction towards the input image. We demonstrate that disentangled controls can be easily achieved in the few-step diffusion model by conditioning on an (automatically generated) detailed text prompt. To manipulate the inverted image, we freeze the noise maps and modify one attribute in the text prompt (either manually or via instruction based editing driven by an LLM), resulting in the generation of a new image similar to the input image with only one attribute changed. It can further control the editing strength and accept instructive text prompt. Our approach facilitates realistic text-guided image edits in real-time, requiring only 8 number of functional evaluations (NFEs) in inversion (one-time cost) and 4 NFEs per edit. Our method is not only fast, but also significantly outperforms state-of-the-art multi-step diffusion editing techniques.

AK

16,062 次观看 • 1 年前

Stable Diffusion 3.5 from Stability AI is now LIVE on Civitai! Stable Diffusion 3.5 Large is a Multimodal Diffusion Transformer text-to-image model that features improved performance in image quality, typography, complex prompt understanding, and resource-efficiency. Generate with SD 3.5 Live on Civitai here👇🏽 All images used to create this video were generated with SD 3.5 by our community 🫶🏽

Stable Diffusion 3.5 from Stability AI is now LIVE on Civitai! Stable Diffusion 3.5 Large is a Multimodal Diffusion Transformer text-to-image model that features improved performance in image quality, typography, complex prompt understanding, and resource-efficiency. Generate with SD 3.5 Live on Civitai here👇🏽 All images used to create this video were generated with SD 3.5 by our community 🫶🏽

Civitai

31,210 次观看 • 1 年前

LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models paper page: github: Recent advancements in text-to-image generation with diffusion models have yielded remarkable results synthesizing highly realistic and diverse images. However, these models still encounter difficulties when generating images from prompts that demand spatial or common sense reasoning. We propose to equip diffusion models with enhanced reasoning capabilities by using off-the-shelf pretrained large language models (LLMs) in a novel two-stage generation process. First, we adapt an LLM to be a text-guided layout generator through in-context learning. When provided with an image prompt, an LLM outputs a scene layout in the form of bounding boxes along with corresponding individual descriptions. Second, we steer a diffusion model with a novel controller to generate images conditioned on the layout. Both stages utilize frozen pretrained models without any LLM or diffusion model parameter optimization. We validate the superiority of our design by demonstrating its ability to outperform the base diffusion model in accurately generating images according to prompts that necessitate both language and spatial reasoning. Additionally, our method naturally allows dialog-based scene specification and is able to handle prompts in a language that is not well-supported by the underlying diffusion model.

LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models paper page: github: Recent advancements in text-to-image generation with diffusion models have yielded remarkable results synthesizing highly realistic and diverse images. However, these models still encounter difficulties when generating images from prompts that demand spatial or common sense reasoning. We propose to equip diffusion models with enhanced reasoning capabilities by using off-the-shelf pretrained large language models (LLMs) in a novel two-stage generation process. First, we adapt an LLM to be a text-guided layout generator through in-context learning. When provided with an image prompt, an LLM outputs a scene layout in the form of bounding boxes along with corresponding individual descriptions. Second, we steer a diffusion model with a novel controller to generate images conditioned on the layout. Both stages utilize frozen pretrained models without any LLM or diffusion model parameter optimization. We validate the superiority of our design by demonstrating its ability to outperform the base diffusion model in accurately generating images according to prompts that necessitate both language and spatial reasoning. Additionally, our method naturally allows dialog-based scene specification and is able to handle prompts in a language that is not well-supported by the underlying diffusion model.

AK

83,657 次观看 • 2 年前

9 Key AI Concepts Explained in 7 minutes - Tokenization - Text Decoding - Prompt Engineering - Multi Step AI Agents - RAGs - RLHF - VAE - Diffusion Models - LoRA

9 Key AI Concepts Explained in 7 minutes - Tokenization - Text Decoding - Prompt Engineering - Multi Step AI Agents - RAGs - RLHF - VAE - Diffusion Models - LoRA

Bytebytego

98,330 次观看 • 3 个月前

In Prompt Engineering for Vision Models, taught by Abby Jacques Verre and Caleb Kaiser of Comet , you’ll learn how to prompt and fine-tune vision models for personalized image generation, image editing, object detection and segmentation. The prompts you'll use for vision models could be text, point coordinates, or bounding boxes, depending on the model. You'll also learn to tune hyperparameters to shape the output. Models you'll use include Segment-Anything Model (SAM), OWL-ViT, and Stable Diffusion. You'll also learn to fine-tune Stable Diffusion to generate personalized images (say, an image of a specific person), using a handful of images for training. As an example of a multi-step workflow, you'll use OWL-ViT to detect an object based on a text prompt, then pass the bounding box to SAM to create a segmentation mask, and input that mask into Stable Diffusion to replace the original object with a new one based on a text prompt. Controlling vision models can be tricky; this course will teach prompting and fine-tuning techniques to get precise control over their output. Get started here:

In Prompt Engineering for Vision Models, taught by Abby Jacques Verre and Caleb Kaiser of Comet , you’ll learn how to prompt and fine-tune vision models for personalized image generation, image editing, object detection and segmentation. The prompts you'll use for vision models could be text, point coordinates, or bounding boxes, depending on the model. You'll also learn to tune hyperparameters to shape the output. Models you'll use include Segment-Anything Model (SAM), OWL-ViT, and Stable Diffusion. You'll also learn to fine-tune Stable Diffusion to generate personalized images (say, an image of a specific person), using a handful of images for training. As an example of a multi-step workflow, you'll use OWL-ViT to detect an object based on a text prompt, then pass the bounding box to SAM to create a segmentation mask, and input that mask into Stable Diffusion to replace the original object with a new one based on a text prompt. Controlling vision models can be tricky; this course will teach prompting and fine-tuning techniques to get precise control over their output. Get started here:

Andrew Ng

151,198 次观看 • 2 年前

Announcing Diffusion Forcing Transformer (DFoT), our new video diffusion algorithm that generates ultra-long videos of 800+ frames. DFoT enables History Guidance, a simple add-on to any existing video diffusion models for a quality boost. Website: (1/7)

Announcing Diffusion Forcing Transformer (DFoT), our new video diffusion algorithm that generates ultra-long videos of 800+ frames. DFoT enables History Guidance, a simple add-on to any existing video diffusion models for a quality boost. Website: (1/7)

Boyuan Chen

175,996 次观看 • 1 年前

📢 IntrinsiX: High-Quality PBR Generation using Image Priors 📢 From text input, we generate renderable PBR maps! Next to editable image generation, our predictions can be distilled into room-scale scenes using SDS for large-scale PBR texture generation. We first train separate LoRA modules for the intrinsic properties of albedo, rough/metal, normal. Then, we introduce cross-intrinsic attention using a rerendering loss with importance-weighted light sampling to enable coherent PBR generation. Our method outperforms text -> image -> PBR methods both in generalization and quality, since directly generating PBR maps does not suffer from the inherent ambiguity of intrinsic image decomposition. In addition, our design choice facilitates SDS-based PBR texture distillation. 🌍 🎥 Great work by Peter Kocsis, Lukas Höllein

📢 IntrinsiX: High-Quality PBR Generation using Image Priors 📢 From text input, we generate renderable PBR maps! Next to editable image generation, our predictions can be distilled into room-scale scenes using SDS for large-scale PBR texture generation. We first train separate LoRA modules for the intrinsic properties of albedo, rough/metal, normal. Then, we introduce cross-intrinsic attention using a rerendering loss with importance-weighted light sampling to enable coherent PBR generation. Our method outperforms text -> image -> PBR methods both in generalization and quality, since directly generating PBR maps does not suffer from the inherent ambiguity of intrinsic image decomposition. In addition, our design choice facilitates SDS-based PBR texture distillation. 🌍 🎥 Great work by Peter Kocsis, Lukas Höllein

Matthias Niessner

21,874 次观看 • 1 年前

Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models turn the publicly available, state-of-the-art text-to-image LDM Stable Diffusion into an efficient and expressive text-to-video model with resolution up to 1280 x 2048 abs: project page:

Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models turn the publicly available, state-of-the-art text-to-image LDM Stable Diffusion into an efficient and expressive text-to-video model with resolution up to 1280 x 2048 abs: project page:

AK

718,682 次观看 • 3 年前

Nvidia presents Edify 3D! The method can generate high-quality 3D assets from text descriptions. It uses a diffusion model to create detailed quad-mesh topologies and high-resolution textures in under 2 minutes.

Nvidia presents Edify 3D! The method can generate high-quality 3D assets from text descriptions. It uses a diffusion model to create detailed quad-mesh topologies and high-resolution textures in under 2 minutes.

Dreaming Tulpa 🥓👑

39,338 次观看 • 1 年前

V3D Video Diffusion Models are Effective 3D Generators Automatic 3D generation has recently attracted widespread attention. Recent methods have greatly accelerated the generation speed, but usually produce less-detailed objects due to limited model capacity or 3D data. Motivated by recent advancements in video diffusion models, we introduce V3D, which leverages the world simulation capacity of pre-trained video diffusion models to facilitate 3D generation. To fully unleash the potential of video diffusion to perceive the 3D world, we further introduce geometrical consistency prior and extend the video diffusion model to a multi-view consistent 3D generator. Benefiting from this, the state-of-the-art video diffusion model could be fine-tuned to generate 360degree orbit frames surrounding an object given a single image. With our tailored reconstruction pipelines, we can generate high-quality meshes or 3D Gaussians within 3 minutes. Furthermore, our method can be extended to scene-level novel view synthesis, achieving precise control over the camera path with sparse input views. Extensive experiments demonstrate the superior performance of the proposed approach, especially in terms of generation quality and multi-view consistency

V3D Video Diffusion Models are Effective 3D Generators Automatic 3D generation has recently attracted widespread attention. Recent methods have greatly accelerated the generation speed, but usually produce less-detailed objects due to limited model capacity or 3D data. Motivated by recent advancements in video diffusion models, we introduce V3D, which leverages the world simulation capacity of pre-trained video diffusion models to facilitate 3D generation. To fully unleash the potential of video diffusion to perceive the 3D world, we further introduce geometrical consistency prior and extend the video diffusion model to a multi-view consistent 3D generator. Benefiting from this, the state-of-the-art video diffusion model could be fine-tuned to generate 360degree orbit frames surrounding an object given a single image. With our tailored reconstruction pipelines, we can generate high-quality meshes or 3D Gaussians within 3 minutes. Furthermore, our method can be extended to scene-level novel view synthesis, achieving precise control over the camera path with sparse input views. Extensive experiments demonstrate the superior performance of the proposed approach, especially in terms of generation quality and multi-view consistency

AK

31,997 次观看 • 2 年前

Diffusion models generate high-quality images but require hundreds of forward passes. MIT CSAIL and Adobe Research introduce Distribution Matching Distillation (DMD), a distillation approach that converts costly multi-step diffusion models into fast one-step generators. A thread 🧵

Diffusion models generate high-quality images but require hundreds of forward passes. MIT CSAIL and Adobe Research introduce Distribution Matching Distillation (DMD), a distillation approach that converts costly multi-step diffusion models into fast one-step generators. A thread 🧵

MIT CSAIL

34,347 次观看 • 2 年前

Introducing the Ad Concepter App. A really simple way to explore new concepts, compositions and story beats for your ads with a single prompt, reference image and product shot. Available now on the web app. Get started at the link below.

Introducing the Ad Concepter App. A really simple way to explore new concepts, compositions and story beats for your ads with a single prompt, reference image and product shot. Available now on the web app. Get started at the link below.

Runway

60,389 次观看 • 2 个月前

MusicHiFi Fast High-Fidelity Stereo Vocoding Diffusion-based audio and music generation models commonly generate music by constructing an image representation of audio (e.g., a mel-spectrogram) and then converting it to audio using a phase reconstruction model or vocoder.

MusicHiFi Fast High-Fidelity Stereo Vocoding Diffusion-based audio and music generation models commonly generate music by constructing an image representation of audio (e.g., a mel-spectrogram) and then converting it to audio using a phase reconstruction model or vocoder.

AK

27,285 次观看 • 2 年前

StyleDrop: Text-to-Image Generation in Any Style introduce StyleDrop, a method that enables the synthesis of images that faithfully follow a specific style using a text-to-image model. The proposed method is extremely versatile and captures nuances and details of a user-provided style, such as color schemes, shading, design patterns, and local and global effects. It efficiently learns a new style by fine-tuning very few trainable parameters (less than 1% of total model parameters) and improving the quality via iterative training with either human or automated feedback. Better yet, StyleDrop is able to deliver impressive results even when the user supplies only a single image that specifies the desired style. An extensive study shows that, for the task of style tuning text-to-image models, StyleDrop implemented on Muse convincingly outperforms other methods, including DreamBooth and textual inversion on Imagen or Stable Diffusion. paper page:

StyleDrop: Text-to-Image Generation in Any Style introduce StyleDrop, a method that enables the synthesis of images that faithfully follow a specific style using a text-to-image model. The proposed method is extremely versatile and captures nuances and details of a user-provided style, such as color schemes, shading, design patterns, and local and global effects. It efficiently learns a new style by fine-tuning very few trainable parameters (less than 1% of total model parameters) and improving the quality via iterative training with either human or automated feedback. Better yet, StyleDrop is able to deliver impressive results even when the user supplies only a single image that specifies the desired style. An extensive study shows that, for the task of style tuning text-to-image models, StyleDrop implemented on Muse convincingly outperforms other methods, including DreamBooth and textual inversion on Imagen or Stable Diffusion. paper page:

AK

56,357 次观看 • 3 年前