Attention Calibration for Disentangled Text-to-Image Personalization

• URL: http://arxiv.org/abs/2403.18551v2
• Date: Thu, 11 Apr 2024 14:10:43 GMT
• Title: Attention Calibration for Disentangled Text-to-Image Personalization
• Authors: Yanbing Zhang, Mengping Yang, Qin Zhou, Zhe Wang,
• Abstract summary: We propose an attention calibration mechanism to improve the concept-level understanding of the T2I model. We demonstrate that our method outperforms the current state of the art in both qualitative and quantitative evaluations.
• Score: 12.339742346826403
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent thrilling progress in large-scale text-to-image (T2I) models has unlocked unprecedented synthesis quality of AI-generated content (AIGC) including image generation, 3D and video composition. Further, personalized techniques enable appealing customized production of a novel concept given only several images as reference. However, an intriguing problem persists: Is it possible to capture multiple, novel concepts from one single reference image? In this paper, we identify that existing approaches fail to preserve visual consistency with the reference image and eliminate cross-influence from concepts. To alleviate this, we propose an attention calibration mechanism to improve the concept-level understanding of the T2I model. Specifically, we first introduce new learnable modifiers bound with classes to capture attributes of multiple concepts. Then, the classes are separated and strengthened following the activation of the cross-attention operation, ensuring comprehensive and self-contained concepts. Additionally, we suppress the attention activation of different classes to mitigate mutual influence among concepts. Together, our proposed method, dubbed DisenDiff, can learn disentangled multiple concepts from one single image and produce novel customized images with learned concepts. We demonstrate that our method outperforms the current state of the art in both qualitative and quantitative evaluations. More importantly, our proposed techniques are compatible with LoRA and inpainting pipelines, enabling more interactive experiences.

Related papers

• MagicTailor: Component-Controllable Personalization in Text-to-Image Diffusion Models [51.1034358143232]
  We introduce component-controllable personalization, a novel task that pushes the boundaries of text-to-image (T2I) models. To overcome these challenges, we design MagicTailor, an innovative framework that leverages Dynamic Masked Degradation (DM-Deg) to dynamically perturb undesired visual semantics.
  arXiv  Detail & Related papers  (2024-10-17T09:22:53Z)
• AttenCraft: Attention-guided Disentanglement of Multiple Concepts for Text-to-Image Customization [4.544788024283586]
  AttenCraft is an attention-guided method for multiple concept disentanglement. We introduce Uniform sampling and Reweighted sampling schemes to alleviate the non-synchronicity of feature acquisition from different concepts. Our method outperforms baseline models in terms of image-alignment, and behaves comparably on text-alignment.
  arXiv  Detail & Related papers  (2024-05-28T08:50:14Z)
• Visual Concept-driven Image Generation with Text-to-Image Diffusion Model [65.96212844602866]
  Text-to-image (TTI) models have demonstrated impressive results in generating high-resolution images of complex scenes. Recent approaches have extended these methods with personalization techniques that allow them to integrate user-illustrated concepts. However, the ability to generate images with multiple interacting concepts, such as human subjects, as well as concepts that may be entangled in one, or across multiple, image illustrations remains illusive. We propose a concept-driven TTI personalization framework that addresses these core challenges.
  arXiv  Detail & Related papers  (2024-02-18T07:28:37Z)
• Decoupled Textual Embeddings for Customized Image Generation [62.98933630971543]
  Customized text-to-image generation aims to learn user-specified concepts with a few images. Existing methods usually suffer from overfitting issues and entangle the subject-unrelated information with the learned concept. We propose the DETEX, a novel approach that learns the disentangled concept embedding for flexible customized text-to-image generation.
  arXiv  Detail & Related papers  (2023-12-19T03:32:10Z)
• Multi-Concept T2I-Zero: Tweaking Only The Text Embeddings and Nothing Else [75.6806649860538]
  We consider a more ambitious goal: natural multi-concept generation using a pre-trained diffusion model. We observe concept dominance and non-localized contribution that severely degrade multi-concept generation performance. We design a minimal low-cost solution that overcomes the above issues by tweaking the text embeddings for more realistic multi-concept text-to-image generation.
  arXiv  Detail & Related papers  (2023-10-11T12:05:44Z)
• Create Your World: Lifelong Text-to-Image Diffusion [75.14353789007902]
  We propose Lifelong text-to-image Diffusion Model (L2DM) to overcome knowledge "catastrophic forgetting" for the past encountered concepts. In respect of knowledge "catastrophic forgetting", our L2DM framework devises a task-aware memory enhancement module and a elastic-concept distillation module. Our model can generate more faithful image across a range of continual text prompts in terms of both qualitative and quantitative metrics.
  arXiv  Detail & Related papers  (2023-09-08T16:45:56Z)
• Break-A-Scene: Extracting Multiple Concepts from a Single Image [80.47666266017207]
  We introduce the task of textual scene decomposition. We propose augmenting the input image with masks that indicate the presence of target concepts. We then present a novel two-phase customization process.
  arXiv  Detail & Related papers  (2023-05-25T17:59:04Z)
• Key-Locked Rank One Editing for Text-to-Image Personalization [43.195870616558935]
  We present Perfusion, a T2I personalization method that addresses challenges using dynamic rank-1 updates to the underlying T2I model. Perfusion avoids overfitting by introducing a new mechanism that "locks" new concepts' cross-attention Keys to their superordinate category. We show that Perfusion outperforms strong baselines in both qualitative and quantitative terms.
  arXiv  Detail & Related papers  (2023-05-02T17:56:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.