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。

Can a small academic team build a strong text-to-image model using only public datasets? Introducing i1: a simple, fully open recipe for strong text-to-image models

Zhuang Liu

12,437 subscribers

68,985 次观看 • 27 天前 •via X (Twitter)

教育科学技术

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

Introducing StyleDrop, a model that allows a significantly higher level of stylized text-to-image synthesis by using a few style reference images that describe the style for text-to-image generation, bypassing the burden of text prompt engineering. More→

Introducing StyleDrop, a model that allows a significantly higher level of stylized text-to-image synthesis by using a few style reference images that describe the style for text-to-image generation, bypassing the burden of text prompt engineering. More→

Google AI

80,378 次观看 • 2 年前

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

165,165 次观看 • 3 年前

CosmicMan A Text-to-Image Foundation Model for Humans We present CosmicMan, a text-to-image foundation model specialized for generating high-fidelity human images. Unlike current general-purpose foundation models that are stuck in the dilemma of inferior quality and

CosmicMan A Text-to-Image Foundation Model for Humans We present CosmicMan, a text-to-image foundation model specialized for generating high-fidelity human images. Unlike current general-purpose foundation models that are stuck in the dilemma of inferior quality and

AK

46,778 次观看 • 2 年前

Introducing Modality Forcing, a recipe for post-training T2I models for SOTA RGB-Depth generation! Text-to-image (T2I) models learn rich representations of the spatial world. How do we build on this prior for high-quality depth generation? 🧵 [1/6]

Introducing Modality Forcing, a recipe for post-training T2I models for SOTA RGB-Depth generation! Text-to-image (T2I) models learn rich representations of the spatial world. How do we build on this prior for high-quality depth generation? 🧵 [1/6]

Bardienus Duisterhof

64,543 次观看 • 1 个月前

wow.. text to 3D would AI is here you can generate 3D assets using text/image and build up a 3D world.. and even auto rig characters. link in comment

wow.. text to 3D would AI is here you can generate 3D assets using text/image and build up a 3D world.. and even auto rig characters. link in comment

el.cine

212,681 次观看 • 1 年前

DiffSplat Repurposing Image Diffusion Models for Scalable Gaussian Splat Generation DiffSplat is a generative framework to synthesize 3D Gaussian Splats from text prompts & single-view images in ⚡️ 1~2 seconds. It is fine-tuned directly from a pretrained text-to-image diffusion model.

DiffSplat Repurposing Image Diffusion Models for Scalable Gaussian Splat Generation DiffSplat is a generative framework to synthesize 3D Gaussian Splats from text prompts & single-view images in ⚡️ 1~2 seconds. It is fine-tuned directly from a pretrained text-to-image diffusion model.

AK

38,416 次观看 • 1 年前

New text and image to video generation AI model Open-Sora-Plan-v1.3.0

New text and image to video generation AI model Open-Sora-Plan-v1.3.0

AK

51,843 次观看 • 1 年前

This new AI image generator is impressive. You can now create an image based on a simple sketch and text prompt.

This new AI image generator is impressive. You can now create an image based on a simple sketch and text prompt.

Rowan Cheung

543,208 次观看 • 3 年前

BUILD 🔥: Microsoft is preparing new image and voice models for the announcement on June 2. > MAI Voice 2, a multilingual model supporting 15 news languages and a wider range of emotional spectrum (check voice samples in the article) > MAI Transcribe 1.5, a new model for speech-to-text use cases. > MAI Image 2.5, already announced last week, is now available on LM Arena in preview. Compared to MAI Image 2, it supports file uploads and can be used for image editing.

BUILD 🔥: Microsoft is preparing new image and voice models for the announcement on June 2. > MAI Voice 2, a multilingual model supporting 15 news languages and a wider range of emotional spectrum (check voice samples in the article) > MAI Transcribe 1.5, a new model for speech-to-text use cases. > MAI Image 2.5, already announced last week, is now available on LM Arena in preview. Compared to MAI Image 2, it supports file uploads and can be used for image editing.

🚨 AI News | TestingCatalog

46,535 次观看 • 1 个月前

[CLIP] by Hand ✍️ The CLIP (Contrastive Language–Image Pre-training) model, a groundbreaking work by OpenAI, redefines the intersection of computer vision and natural language processing. It is the basis of all the multi-modal foundation models we see today. How does CLIP work? Goal: 🟨 Learn a shared embedding space for text and image [1] Given ↳ A mini batch of 3 text-image pairs ↳ OpenAI used 400 million text-image pairs to train its original CLIP model. Process 1st pair: "big table" [2] 🟪 Text → 2 Vectors (3D) ↳ Look up word embedding vectors using word2vec. [3] 🟩 Image → 2 Vectors (4D) ↳ Divide the image into two patches. ↳ Flatten each patch [4] Process other pairs ↳ Repeat [2]-[3] [5] 🟪 Text Encoder & 🟩 Image Encoder ↳ Encode input vectors into feature vectors ↳ Here, both encoders are simple one layer perceptron (linear + ReLU) ↳ In practice, the encoders are usually transformer models. [6] 🟪 🟩 Mean Pooling: 2 → 1 vector ↳ Average 2 feature vectors into a single vector by averaging across the columns ↳ The goal is to have one vector to represent each image or text [7] 🟪 🟩 -> 🟨 Projection ↳ Note that the text and image feature vectors from the encoders have different dimensions (3D vs. 4D). ↳ Use a linear layer to project image and text vectors to a 2D shared embedding space. 🏋️ Contrastive Pre-training 🏋️ [8] Prepare for MatMul ↳ Copy text vectors (T1,T2,T3) ↳ Copy the transpose of image vectors (I1,I2,I3) ↳ They are all in the 2D shared embedding space. [9] 🟦 MatMul ↳ Multiply T and I matrices. ↳ This is equivalent to taking dot product between every pair of image and text vectors. ↳ The purpose is to use dot product to estimate the similarity between a pair of image-text. [10] 🟦 Softmax: e^x ↳ Raise e to the power of the number in each cell ↳ To simplify hand calculation, we approximate e^□ with 3^□. [11] 🟦 Softmax: ∑ ↳ Sum each row for 🟩 image→🟪 text ↳ Sum each column for 🟪 text→ 🟩 image [12] 🟦 Softmax: 1 / sum ↳ Divide each element by the column sum to obtain a similarity matrix for 🟪 text→🟩 image ↳ Divide each element by the row sum to obtain a similarity matrix for 🟩 image→🟪 text [13] 🟥 Loss Gradients ↳ The "Targets" for the similarity matrices are Identity Matrices. ↳ Why? If I and T come from the same pair (i=j), we want the highest value, which is 1, and 0 otherwise. ↳ Apply the simple equation of [Similarity - Target] to compute gradients of for both directions. ↳ Why so simple? Because when Softmax and Cross-Entropy Loss are used together, the math magically works out that way. ↳ These gradients kick off the backpropagation process to update weights and biases of the encoders and projection layers (red borders).

[CLIP] by Hand ✍️ The CLIP (Contrastive Language–Image Pre-training) model, a groundbreaking work by OpenAI, redefines the intersection of computer vision and natural language processing. It is the basis of all the multi-modal foundation models we see today. How does CLIP work? Goal: 🟨 Learn a shared embedding space for text and image [1] Given ↳ A mini batch of 3 text-image pairs ↳ OpenAI used 400 million text-image pairs to train its original CLIP model. Process 1st pair: "big table" [2] 🟪 Text → 2 Vectors (3D) ↳ Look up word embedding vectors using word2vec. [3] 🟩 Image → 2 Vectors (4D) ↳ Divide the image into two patches. ↳ Flatten each patch [4] Process other pairs ↳ Repeat [2]-[3] [5] 🟪 Text Encoder & 🟩 Image Encoder ↳ Encode input vectors into feature vectors ↳ Here, both encoders are simple one layer perceptron (linear + ReLU) ↳ In practice, the encoders are usually transformer models. [6] 🟪 🟩 Mean Pooling: 2 → 1 vector ↳ Average 2 feature vectors into a single vector by averaging across the columns ↳ The goal is to have one vector to represent each image or text [7] 🟪 🟩 -> 🟨 Projection ↳ Note that the text and image feature vectors from the encoders have different dimensions (3D vs. 4D). ↳ Use a linear layer to project image and text vectors to a 2D shared embedding space. 🏋️ Contrastive Pre-training 🏋️ [8] Prepare for MatMul ↳ Copy text vectors (T1,T2,T3) ↳ Copy the transpose of image vectors (I1,I2,I3) ↳ They are all in the 2D shared embedding space. [9] 🟦 MatMul ↳ Multiply T and I matrices. ↳ This is equivalent to taking dot product between every pair of image and text vectors. ↳ The purpose is to use dot product to estimate the similarity between a pair of image-text. [10] 🟦 Softmax: e^x ↳ Raise e to the power of the number in each cell ↳ To simplify hand calculation, we approximate e^□ with 3^□. [11] 🟦 Softmax: ∑ ↳ Sum each row for 🟩 image→🟪 text ↳ Sum each column for 🟪 text→ 🟩 image [12] 🟦 Softmax: 1 / sum ↳ Divide each element by the column sum to obtain a similarity matrix for 🟪 text→🟩 image ↳ Divide each element by the row sum to obtain a similarity matrix for 🟩 image→🟪 text [13] 🟥 Loss Gradients ↳ The "Targets" for the similarity matrices are Identity Matrices. ↳ Why? If I and T come from the same pair (i=j), we want the highest value, which is 1, and 0 otherwise. ↳ Apply the simple equation of [Similarity - Target] to compute gradients of for both directions. ↳ Why so simple? Because when Softmax and Cross-Entropy Loss are used together, the math magically works out that way. ↳ These gradients kick off the backpropagation process to update weights and biases of the encoders and projection layers (red borders).

Tom Yeh

67,834 次观看 • 2 年前

📢 A Recipe for Generating 3D Worlds From a Single Image 📢 Our recipe explains how existing generative models can be adapted with minimal training effort to generate 3D worlds from a single input image.

📢 A Recipe for Generating 3D Worlds From a Single Image 📢 Our recipe explains how existing generative models can be adapted with minimal training effort to generate 3D worlds from a single input image.

Katja Schwarz

13,970 次观看 • 1 年前

Using CSS mask-image you can truncate texts with a smooth fade effect. It looks clean and it's a great alternative to text ellipsis. .text-container { overflow: hidden; white-space: nowrap; -webkit-mask-image: linear-gradient( to right, black 50%, transparent); mask-image: linear-gradient( to right, black 50%, transparent); } } If you found this one useful, follow for more! ❤️

Using CSS mask-image you can truncate texts with a smooth fade effect. It looks clean and it's a great alternative to text ellipsis. .text-container { overflow: hidden; white-space: nowrap; -webkit-mask-image: linear-gradient( to right, black 50%, transparent); mask-image: linear-gradient( to right, black 50%, transparent); } } If you found this one useful, follow for more! ❤️

Ali

13,380 次观看 • 1 个月前

Phoenix GenAI’s new generative AI model Image-to-3D is now online. Create high fidelity 3D models importable to Unreal Engine 5 or Unity simply by sending an original image into PhoenixLLM. Better yet, generate the source image using Phoenix GenAI’s Flux and feed it into Image-to-3D. This release marks yet another upgrade of GenAI’s arsenal of capabilities, getting it ready for multi-workflow GenAI agents, in which users will be able to combine text-to-image, image-to-prompt, text-to-video, text-to-3D, and image-to-3D into complex multi-step workflows with simple commands via PhoenixLLM. Image-to-3D is yet another addition to Phoenix’s Vertical AI Solutions for gaming, content, and metaverse. Users are able to use it as a Phoenix-native alternative to SkyNet AI Marketplace’s Tripo Integration earlier this year. #Phoenix $PHB

Phoenix GenAI’s new generative AI model Image-to-3D is now online. Create high fidelity 3D models importable to Unreal Engine 5 or Unity simply by sending an original image into PhoenixLLM. Better yet, generate the source image using Phoenix GenAI’s Flux and feed it into Image-to-3D. This release marks yet another upgrade of GenAI’s arsenal of capabilities, getting it ready for multi-workflow GenAI agents, in which users will be able to combine text-to-image, image-to-prompt, text-to-video, text-to-3D, and image-to-3D into complex multi-step workflows with simple commands via PhoenixLLM. Image-to-3D is yet another addition to Phoenix’s Vertical AI Solutions for gaming, content, and metaverse. Users are able to use it as a Phoenix-native alternative to SkyNet AI Marketplace’s Tripo Integration earlier this year. #Phoenix $PHB

Phoenix AI

30,853 次观看 • 1 年前

🤯 OneDiffusion: A versatile, large-scale diffusion model that seamlessly supports bidirectional image synthesis and understanding across diverse tasks. ✅ Text to Image ✅ Image to Depth ✅ Image to Segmentation ✅ Image to Pose ✅ FaceID ✅ Image to Multiview How to use & more👇

🤯 OneDiffusion: A versatile, large-scale diffusion model that seamlessly supports bidirectional image synthesis and understanding across diverse tasks. ✅ Text to Image ✅ Image to Depth ✅ Image to Segmentation ✅ Image to Pose ✅ FaceID ✅ Image to Multiview How to use & more👇

Gradio

11,820 次观看 • 1 年前

Introducing SDXL Turbo: A real-time text-to-image generation model. SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one. The code, research paper, and weights for non-commercial use are now available on our website. You can test SDXL Turbo on Stability AI’s image editing platform Clipdrop, with a beta demonstration of the real-time text-to-image generation capabilities. Learn more:

Introducing SDXL Turbo: A real-time text-to-image generation model. SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one. The code, research paper, and weights for non-commercial use are now available on our website. You can test SDXL Turbo on Stability AI’s image editing platform Clipdrop, with a beta demonstration of the real-time text-to-image generation capabilities. Learn more:

Stability AI

976,344 次观看 • 2 年前

The Microsoft AI model family expands in Microsoft Foundry with 7 new models. A complete multimodal stack: Text, image, transcription, and voice, ready for developers to build with under one set of governance and security controls. #MSBuild

The Microsoft AI model family expands in Microsoft Foundry with 7 new models. A complete multimodal stack: Text, image, transcription, and voice, ready for developers to build with under one set of governance and security controls. #MSBuild

Microsoft

50,254 次观看 • 1 个月前

Introducing the all new “AI Models” by GPUNET It’s time for something new. We’ve officially launched a brand new line of core AI products: → Powerful LLMs → Text-to-Image gen → Text-to-Video gen And you’re invited to try it. Get Free Credits →

GPU.NET

27,051 次观看 • 1 年前

Introducing Ideogram 1.0: the most advanced text-to-image model, now available on This offers state-of-the-art text rendering, unprecedented photorealism, exceptional prompt adherence, and a new feature called Magic Prompt to help with prompting.

Introducing Ideogram 1.0: the most advanced text-to-image model, now available on This offers state-of-the-art text rendering, unprecedented photorealism, exceptional prompt adherence, and a new feature called Magic Prompt to help with prompting.

Ideogram

256,056 次观看 • 2 年前

We’re excited to announce the release and open-source of HunyuanImage 3.0 — the largest and most powerful open-source text-to-image model to date, with over 80 billion total parameters, of which 13 billion are activated per token during inference.The effect is completely comparable to the industry’s flagship closed-source model.🚀🚀🚀 HunyuanImage 3.0 originates from our internally developed native multimodal large language model, with fine-tuning and post-training focused on text-to-image generation. This unique foundation gives the model a powerful set of capabilities: ✅Reason with world knowledge ✅Understand complex, thousand-word prompts ✅Generate precise text within images Different from traditional DiT architecture image generation models, HunyuanImage 3.0’s MoE architecture uses a Transfusion-based approach to deeply couple Diffusion and LLM training for a single, powerful system. Built on Hunyuan-A13B, HunyuanImage 3.0 was trained on a massive dataset: 5 billion image-text pairs, video frames, interleaved image-text data, and 6 trillion tokens of text corpora. This hybrid training across multimodal generation, understanding, and LLM capabilities allows the model to seamlessly integrate multiple tasks. Whether you're an illustrator, designer, or creator, this is built to slash your workflow from hours to minutes. HunyuanImage 3.0 can generate intricate text, detailed comics, expressive emojis, and lively, engaging illustrations for educational content. The current release focuses solely on text-to-image generation and future updates will include image-to-image, image editing, multi-turn interaction, and more. 👉🏻Try it now: 🔗GitHub: 🤗Hugging Face:

We’re excited to announce the release and open-source of HunyuanImage 3.0 — the largest and most powerful open-source text-to-image model to date, with over 80 billion total parameters, of which 13 billion are activated per token during inference.The effect is completely comparable to the industry’s flagship closed-source model.🚀🚀🚀 HunyuanImage 3.0 originates from our internally developed native multimodal large language model, with fine-tuning and post-training focused on text-to-image generation. This unique foundation gives the model a powerful set of capabilities: ✅Reason with world knowledge ✅Understand complex, thousand-word prompts ✅Generate precise text within images Different from traditional DiT architecture image generation models, HunyuanImage 3.0’s MoE architecture uses a Transfusion-based approach to deeply couple Diffusion and LLM training for a single, powerful system. Built on Hunyuan-A13B, HunyuanImage 3.0 was trained on a massive dataset: 5 billion image-text pairs, video frames, interleaved image-text data, and 6 trillion tokens of text corpora. This hybrid training across multimodal generation, understanding, and LLM capabilities allows the model to seamlessly integrate multiple tasks. Whether you're an illustrator, designer, or creator, this is built to slash your workflow from hours to minutes. HunyuanImage 3.0 can generate intricate text, detailed comics, expressive emojis, and lively, engaging illustrations for educational content. The current release focuses solely on text-to-image generation and future updates will include image-to-image, image editing, multi-turn interaction, and more. 👉🏻Try it now: 🔗GitHub: 🤗Hugging Face:

Tencent Hy

412,658 次观看 • 9 个月前

🚀 LongCat-Video Now Open-Source: Text/Image-to-Video + Video Continuation in One Model 🏆 Text/Image-to-Video Performance Hits Open-Source SOTA 🎬 Minutes-Long High-Quality Videos: No Color Drift/Quality Loss (Industry-Standout) ⚙ 13.6B Params | Strong Open-Source DiT-Based Unified Multitask Video Base Model ⚡ C2F Pipeline + Block Sparse Attention: 720p/30fps Video in Minutes 🤗 Open-Source Links: GitHub： Hugging Face： Project Page：

🚀 LongCat-Video Now Open-Source: Text/Image-to-Video + Video Continuation in One Model 🏆 Text/Image-to-Video Performance Hits Open-Source SOTA 🎬 Minutes-Long High-Quality Videos: No Color Drift/Quality Loss (Industry-Standout) ⚙ 13.6B Params | Strong Open-Source DiT-Based Unified Multitask Video Base Model ⚡ C2F Pipeline + Block Sparse Attention: 720p/30fps Video in Minutes 🤗 Open-Source Links: GitHub： Hugging Face： Project Page：

Meituan LongCat

43,748 次观看 • 8 个月前