How Control Information Influences Multilingual Text Image Generation and Editing?
- URL: http://arxiv.org/abs/2407.11502v2
- Date: Sun, 21 Jul 2024 08:22:18 GMT
- Title: How Control Information Influences Multilingual Text Image Generation and Editing?
- Authors: Boqiang Zhang, Zuan Gao, Yadong Qu, Hongtao Xie,
- Abstract summary: We investigate the role of control information in generating high-quality text.
We propose TextGen, a novel framework designed to enhance generation quality by optimizing control information.
Our method achieves state-of-the-art performance in both Chinese and English text generation.
- Score: 28.999640376365335
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Visual text generation has significantly advanced through diffusion models aimed at producing images with readable and realistic text. Recent works primarily use a ControlNet-based framework, employing standard font text images to control diffusion models. Recognizing the critical role of control information in generating high-quality text, we investigate its influence from three perspectives: input encoding, role at different stages, and output features. Our findings reveal that: 1) Input control information has unique characteristics compared to conventional inputs like Canny edges and depth maps. 2) Control information plays distinct roles at different stages of the denoising process. 3) Output control features significantly differ from the base and skip features of the U-Net decoder in the frequency domain. Based on these insights, we propose TextGen, a novel framework designed to enhance generation quality by optimizing control information. We improve input and output features using Fourier analysis to emphasize relevant information and reduce noise. Additionally, we employ a two-stage generation framework to align the different roles of control information at different stages. Furthermore, we introduce an effective and lightweight dataset for training. Our method achieves state-of-the-art performance in both Chinese and English text generation. The code and dataset available at https://github.com/CyrilSterling/TextGen.
Related papers
- DynamiCtrl: Rethinking the Basic Structure and the Role of Text for High-quality Human Image Animation [63.781450025764904]
DynamiCtrl is a novel framework that explores different pose-guided structures in MM-DiT.
We propose Pose-adaptive Layer Norm (PadaLN), which utilizes adaptive layer normalization to encode sparse pose features.
By leveraging text, we not only enable fine-grained control over the generated content, but also, for the first time, achieve simultaneous control over both background and motion.
arXiv Detail & Related papers (2025-03-27T08:07:45Z) - The Narrow Gate: Localized Image-Text Communication in Vision-Language Models [36.33608889682152]
This study investigates how vision-language models handle image-understanding tasks.
We find that in models with multimodal outputs, image and text embeddings are more separated within the residual stream.
In contrast, models trained for image and text generation tend to rely on a single token that acts as a narrow gate for visual information.
arXiv Detail & Related papers (2024-12-09T16:39:40Z) - FBSDiff: Plug-and-Play Frequency Band Substitution of Diffusion Features for Highly Controllable Text-Driven Image Translation [19.65838242227773]
This paper contributes a novel, concise, and efficient approach that adapts pre-trained large-scale text-to-image (T2I) diffusion model to the image-to-image (I2I) paradigm in a plug-and-play manner.
Our method allows flexible control over both guiding factor and guiding intensity of the reference image simply by tuning the type and bandwidth of the substituted frequency band.
arXiv Detail & Related papers (2024-08-02T04:13:38Z) - Contrastive Prompts Improve Disentanglement in Text-to-Image Diffusion
Models [68.47333676663312]
We show a simple modification of classifier-free guidance can help disentangle image factors in text-to-image models.
The key idea of our method, Contrastive Guidance, is to characterize an intended factor with two prompts that differ in minimal tokens.
We illustrate whose benefits in three scenarios: (1) to guide domain-specific diffusion models trained on an object class, (2) to gain continuous, rig-like controls for text-to-image generation, and (3) to improve the performance of zero-shot image editors.
arXiv Detail & Related papers (2024-02-21T03:01:17Z) - Text2Data: Low-Resource Data Generation with Textual Control [104.38011760992637]
Natural language serves as a common and straightforward control signal for humans to interact seamlessly with machines.
We propose Text2Data, a novel approach that utilizes unlabeled data to understand the underlying data distribution through an unsupervised diffusion model.
It undergoes controllable finetuning via a novel constraint optimization-based learning objective that ensures controllability and effectively counteracts catastrophic forgetting.
arXiv Detail & Related papers (2024-02-08T03:41:39Z) - UDiffText: A Unified Framework for High-quality Text Synthesis in
Arbitrary Images via Character-aware Diffusion Models [25.219960711604728]
This paper proposes a novel approach for text image generation, utilizing a pre-trained diffusion model.
Our approach involves the design and training of a light-weight character-level text encoder, which replaces the original CLIP encoder.
By employing an inference stage refinement process, we achieve a notably high sequence accuracy when synthesizing text in arbitrarily given images.
arXiv Detail & Related papers (2023-12-08T07:47:46Z) - Fine-grained Controllable Video Generation via Object Appearance and
Context [74.23066823064575]
We propose fine-grained controllable video generation (FACTOR) to achieve detailed control.
FACTOR aims to control objects' appearances and context, including their location and category.
Our method achieves controllability of object appearances without finetuning, which reduces the per-subject optimization efforts for the users.
arXiv Detail & Related papers (2023-12-05T17:47:33Z) - Controlled Text Generation with Hidden Representation Transformations [12.576140288264835]
CHRT steers large language models to generate text pertaining to certain attributes (such as toxicity)
We employ a contrastive-learning framework to learn these transformations that can be combined to gain multi-attribute control.
CHRT outperforms all the baselines in the task of detoxification, positive sentiment steering, and text simplification.
arXiv Detail & Related papers (2023-05-30T17:21:17Z) - SpaText: Spatio-Textual Representation for Controllable Image Generation [61.89548017729586]
SpaText is a new method for text-to-image generation using open-vocabulary scene control.
In addition to a global text prompt that describes the entire scene, the user provides a segmentation map.
We show its effectiveness on two state-of-the-art diffusion models: pixel-based and latent-conditional-based.
arXiv Detail & Related papers (2022-11-25T18:59:10Z) - FAST: Improving Controllability for Text Generation with Feedback Aware
Self-Training [25.75982440355576]
Controllable text generation systems often leverage control codes to direct various properties of the output like style and length.
Inspired by recent work on causal inference for NLP, this paper reveals a previously overlooked flaw in these control code-based conditional text generation algorithms.
We propose two simple techniques to reduce these correlations in training sets.
arXiv Detail & Related papers (2022-10-06T19:00:51Z) - Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors [58.71128866226768]
Recent text-to-image generation methods have incrementally improved the generated image fidelity and text relevancy.
We propose a novel text-to-image method that addresses these gaps by (i) enabling a simple control mechanism complementary to text in the form of a scene.
Our model achieves state-of-the-art FID and human evaluation results, unlocking the ability to generate high fidelity images in a resolution of 512x512 pixels.
arXiv Detail & Related papers (2022-03-24T15:44:50Z) - Integrating Visuospatial, Linguistic and Commonsense Structure into
Story Visualization [81.26077816854449]
We first explore the use of constituency parse trees for encoding structured input.
Second, we augment the structured input with commonsense information and study the impact of this external knowledge on the generation of visual story.
Third, we incorporate visual structure via bounding boxes and dense captioning to provide feedback about the characters/objects in generated images.
arXiv Detail & Related papers (2021-10-21T00:16:02Z)
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.