Yuanzhi Zhu

Zixiang Zhao, Haowen Bai, Yuanzhi Zhu et al. · eth-zurich

Multi-modality image fusion aims to combine different modalities to produce fused images that retain the complementary features of each modality, such as functional highlights and texture details. To leverage strong generative priors and address challenges such as unstable training and lack of interpretability for GAN-based generative methods, we propose a novel fusion algorithm based on the denoising diffusion probabilistic model (DDPM). The fusion task is formulated as a conditional generation problem under the DDPM sampling framework, which is further divided into an unconditional generation subproblem and a maximum likelihood subproblem. The latter is modeled in a hierarchical Bayesian manner with latent variables and inferred by the expectation-maximization (EM) algorithm. By integrating the inference solution into the diffusion sampling iteration, our method can generate high-quality fused images with natural image generative priors and cross-modality information from source images. Note that all we required is an unconditional pre-trained generative model, and no fine-tuning is needed. Our extensive experiments indicate that our approach yields promising fusion results in infrared-visible image fusion and medical image fusion. The code is available at \url{https://github.com/Zhaozixiang1228/MMIF-DDFM}.

Yuanzhi Zhu, Zhaohai Li, Tianwei Wang et al.

Current text recognition systems, including those for handwritten scripts and scene text, have relied heavily on image synthesis and augmentation, since it is difficult to realize real-world complexity and diversity through collecting and annotating enough real text images. In this paper, we explore the problem of text image generation, by taking advantage of the powerful abilities of Diffusion Models in generating photo-realistic and diverse image samples with given conditions, and propose a method called Conditional Text Image Generation with Diffusion Models (CTIG-DM for short). To conform to the characteristics of text images, we devise three conditions: image condition, text condition, and style condition, which can be used to control the attributes, contents, and styles of the samples in the image generation process. Specifically, four text image generation modes, namely: (1) synthesis mode, (2) augmentation mode, (3) recovery mode, and (4) imitation mode, can be derived by combining and configuring these three conditions. Extensive experiments on both handwritten and scene text demonstrate that the proposed CTIG-DM is able to produce image samples that simulate real-world complexity and diversity, and thus can boost the performance of existing text recognizers. Besides, CTIG-DM shows its appealing potential in domain adaptation and generating images containing Out-Of-Vocabulary (OOV) words.

Yuanzhi Zhu, Xingchao Liu, Qiang Liu

Diffusion models excel in high-quality generation but suffer from slow inference due to iterative sampling. While recent methods have successfully transformed diffusion models into one-step generators, they neglect model size reduction, limiting their applicability in compute-constrained scenarios. This paper aims to develop small, efficient one-step diffusion models based on the powerful rectified flow framework, by exploring joint compression of inference steps and model size. The rectified flow framework trains one-step generative models using two operations, reflow and distillation. Compared with the original framework, squeezing the model size brings two new challenges: (1) the initialization mismatch between large teachers and small students during reflow; (2) the underperformance of naive distillation on small student models. To overcome these issues, we propose Annealing Reflow and Flow-Guided Distillation, which together comprise our SlimFlow framework. With our novel framework, we train a one-step diffusion model with an FID of 5.02 and 15.7M parameters, outperforming the previous state-of-the-art one-step diffusion model (FID=6.47, 19.4M parameters) on CIFAR10. On ImageNet 64$\times$64 and FFHQ 64$\times$64, our method yields small one-step diffusion models that are comparable to larger models, showcasing the effectiveness of our method in creating compact, efficient one-step diffusion models.

Yuanzhi Zhu, Jiawei Liu, Feiyu Gao et al.

Recently, with the rapid advancements of generative models, the field of visual text generation has witnessed significant progress. However, it is still challenging to render high-quality text images in real-world scenarios, as three critical criteria should be satisfied: (1) Fidelity: the generated text images should be photo-realistic and the contents are expected to be the same as specified in the given conditions; (2) Reasonability: the regions and contents of the generated text should cohere with the scene; (3) Utility: the generated text images can facilitate related tasks (e.g., text detection and recognition). Upon investigation, we find that existing methods, either rendering-based or diffusion-based, can hardly meet all these aspects simultaneously, limiting their application range. Therefore, we propose in this paper a visual text generator (termed SceneVTG), which can produce high-quality text images in the wild. Following a two-stage paradigm, SceneVTG leverages a Multimodal Large Language Model to recommend reasonable text regions and contents across multiple scales and levels, which are used by a conditional diffusion model as conditions to generate text images. Extensive experiments demonstrate that the proposed SceneVTG significantly outperforms traditional rendering-based methods and recent diffusion-based methods in terms of fidelity and reasonability. Besides, the generated images provide superior utility for tasks involving text detection and text recognition. Code and datasets are available at AdvancedLiterateMachinery.

Shilin Lu, Zihan Zhou, Jiayou Lu et al.

Current image watermarking methods are vulnerable to advanced image editing techniques enabled by large-scale text-to-image models. These models can distort embedded watermarks during editing, posing significant challenges to copyright protection. In this work, we introduce W-Bench, the first comprehensive benchmark designed to evaluate the robustness of watermarking methods against a wide range of image editing techniques, including image regeneration, global editing, local editing, and image-to-video generation. Through extensive evaluations of eleven representative watermarking methods against prevalent editing techniques, we demonstrate that most methods fail to detect watermarks after such edits. To address this limitation, we propose VINE, a watermarking method that significantly enhances robustness against various image editing techniques while maintaining high image quality. Our approach involves two key innovations: (1) we analyze the frequency characteristics of image editing and identify that blurring distortions exhibit similar frequency properties, which allows us to use them as surrogate attacks during training to bolster watermark robustness; (2) we leverage a large-scale pretrained diffusion model SDXL-Turbo, adapting it for the watermarking task to achieve more imperceptible and robust watermark embedding. Experimental results show that our method achieves outstanding watermarking performance under various image editing techniques, outperforming existing methods in both image quality and robustness. Code is available at https://github.com/Shilin-LU/VINE.

Yuanzhi Zhu, Ruiqing Wang, Shilin Lu et al.

Recent advances in diffusion and flow-based generative models have demonstrated remarkable success in image restoration tasks, achieving superior perceptual quality compared to traditional deep learning approaches. However, these methods either require numerous sampling steps to generate high-quality images, resulting in significant computational overhead, or rely on common model distillation, which usually imposes a fixed fidelity-realism trade-off and thus lacks flexibility. In this paper, we introduce OFTSR, a novel flow-based framework for one-step image super-resolution that can produce outputs with tunable levels of fidelity and realism. Our approach first trains a conditional flow-based super-resolution model to serve as a teacher model. We then distill this teacher model by applying a specialized constraint. Specifically, we force the predictions from our one-step student model for same input to lie on the same sampling ODE trajectory of the teacher model. This alignment ensures that the student model's single-step predictions from initial states match the teacher's predictions from a closer intermediate state. Through extensive experiments on datasets including FFHQ (256$\times$256), DIV2K, and ImageNet (256$\times$256), we demonstrate that OFTSR achieves state-of-the-art performance for one-step image super-resolution, while having the ability to flexibly tune the fidelity-realism trade-off. Codes: \href{https://github.com/yuanzhi-zhu/OFTSR}{https://github.com/yuanzhi-zhu/OFTSR}.

Yuanzhi Zhu, Xi Wang, Stéphane Lathuilière et al.

Diffusion and flow models achieve State-Of-The-Art (SOTA) generative performance, yet many practically important behaviors such as fine-grained prompt fidelity, compositional correctness, and text rendering are weakly specified by score or flow matching pretraining objectives. Reinforcement Learning (RL) fine-tuning with external, black-box rewards is a natural remedy, but diffusion RL is often brittle. Trajectory-based methods incur high memory cost and high-variance gradient estimates; forward-process approaches converge faster but can suffer from distribution drift, and hence reward hacking. In this work, we present \textbf{Centered Reward Distillation (CRD)}, a diffusion RL framework derived from KL-regularized reward maximization built on forward-process-based fine-tuning. The key insight is that the intractable normalizing constant cancels under \emph{within-prompt centering}, yielding a well-posed reward-matching objective. To enable reliable text-to-image fine-tuning, we introduce techniques that explicitly control distribution drift: (\textit{i}) decoupling the sampler from the moving reference to prevent ratio-signal collapse, (\textit{ii}) KL anchoring to a CFG-guided pretrained model to control long-run drift and align with the inference-time semantics of the pre-trained model, and (\textit{iii}) reward-adaptive KL strength to accelerate early learning under large KL regularization while reducing late-stage exploitation of reward-model loopholes. Experiments on text-to-image post-training with \texttt{GenEval} and \texttt{OCR} rewards show that CRD achieves competitive SOTA reward optimization results with fast convergence and reduced reward hacking, as validated on unseen preference metrics.

Shuai Bai, Keqin Chen, Xuejing Liu et al. · pku

We introduce Qwen2.5-VL, the latest flagship model of Qwen vision-language series, which demonstrates significant advancements in both foundational capabilities and innovative functionalities. Qwen2.5-VL achieves a major leap forward in understanding and interacting with the world through enhanced visual recognition, precise object localization, robust document parsing, and long-video comprehension. A standout feature of Qwen2.5-VL is its ability to localize objects using bounding boxes or points accurately. It provides robust structured data extraction from invoices, forms, and tables, as well as detailed analysis of charts, diagrams, and layouts. To handle complex inputs, Qwen2.5-VL introduces dynamic resolution processing and absolute time encoding, enabling it to process images of varying sizes and videos of extended durations (up to hours) with second-level event localization. This allows the model to natively perceive spatial scales and temporal dynamics without relying on traditional normalization techniques. By training a native dynamic-resolution Vision Transformer (ViT) from scratch and incorporating Window Attention, we reduce computational overhead while maintaining native resolution. As a result, Qwen2.5-VL excels not only in static image and document understanding but also as an interactive visual agent capable of reasoning, tool usage, and task execution in real-world scenarios such as operating computers and mobile devices. Qwen2.5-VL is available in three sizes, addressing diverse use cases from edge AI to high-performance computing. The flagship Qwen2.5-VL-72B model matches state-of-the-art models like GPT-4o and Claude 3.5 Sonnet, particularly excelling in document and diagram understanding. Additionally, Qwen2.5-VL maintains robust linguistic performance, preserving the core language competencies of the Qwen2.5 LLM.

Ruiqing Wang, Kai Zhang, Yuanzhi Zhu et al.

Diffusion- and flow-based models have advanced Real-world Image Super-Resolution (Real-ISR), but their multi-step sampling makes inference slow and hard to deploy. One-step distillation alleviates the cost, yet often degrades restoration quality and removes the option to refine with more steps. We present Mean Flows for Super-Resolution (MFSR), a new distillation framework that produces photorealistic results in a single step while still allowing an optional few-step path for further improvement. Our approach uses MeanFlow as the learning target, enabling the student to approximate the average velocity between arbitrary states of the Probability Flow ODE (PF-ODE) and effectively capture the teacher's dynamics without explicit rollouts. To better leverage pretrained generative priors, we additionally improve original MeanFlow's Classifier-Free Guidance (CFG) formulation with teacher CFG distillation strategy, which enhances restoration capability and preserves fine details. Experiments on both synthetic and real-world benchmarks demonstrate that MFSR achieves efficient, flexible, and high-quality super-resolution, delivering results on par with or even better than multi-step teachers while requiring much lower computational cost.

Boqi Chen, Yuanzhi Zhu, Yunke Ao et al.

Single-source domain generalization (SDG) aims to learn a model from a single source domain that can generalize well on unseen target domains. This is an important task in computer vision, particularly relevant to medical imaging where domain shifts are common. In this work, we consider a challenging yet practical setting: SDG for cross-modality medical image segmentation. We combine causality-inspired theoretical insights on learning domain-invariant representations with recent advancements in diffusion-based augmentation to improve generalization across diverse imaging modalities. Guided by the ``intervention-augmentation equivariant'' principle, we use controlled diffusion models (DMs) to simulate diverse imaging styles while preserving the content, leveraging rich generative priors in large-scale pretrained DMs to comprehensively perturb the multidimensional style variable. Extensive experiments on challenging cross-modality segmentation tasks demonstrate that our approach consistently outperforms state-of-the-art SDG methods across three distinct anatomies and imaging modalities. The source code is available at \href{https://github.com/ratschlab/ICMSeg}{https://github.com/ratschlab/ICMSeg}.

Yuanzhi Zhu, Kai Zhang, Jingyun Liang et al.

Plug-and-play Image Restoration (IR) has been widely recognized as a flexible and interpretable method for solving various inverse problems by utilizing any off-the-shelf denoiser as the implicit image prior. However, most existing methods focus on discriminative Gaussian denoisers. Although diffusion models have shown impressive performance for high-quality image synthesis, their potential to serve as a generative denoiser prior to the plug-and-play IR methods remains to be further explored. While several other attempts have been made to adopt diffusion models for image restoration, they either fail to achieve satisfactory results or typically require an unacceptable number of Neural Function Evaluations (NFEs) during inference. This paper proposes DiffPIR, which integrates the traditional plug-and-play method into the diffusion sampling framework. Compared to plug-and-play IR methods that rely on discriminative Gaussian denoisers, DiffPIR is expected to inherit the generative ability of diffusion models. Experimental results on three representative IR tasks, including super-resolution, image deblurring, and inpainting, demonstrate that DiffPIR achieves state-of-the-art performance on both the FFHQ and ImageNet datasets in terms of reconstruction faithfulness and perceptual quality with no more than 100 NFEs. The source code is available at {\url{https://github.com/yuanzhi-zhu/DiffPIR}}

Tianwei Wang, Yuanzhi Zhu, Lianwen Jin et al.

Text recognition is a popular research subject with many associated challenges. Despite the considerable progress made in recent years, the text recognition task itself is still constrained to solve the problem of reading cropped line text images and serves as a subtask of optical character recognition (OCR) systems. As a result, the final text recognition result is limited by the performance of the text detector. In this paper, we propose a simple, elegant and effective paradigm called Implicit Feature Alignment (IFA), which can be easily integrated into current text recognizers, resulting in a novel inference mechanism called IFAinference. This enables an ordinary text recognizer to process multi-line text such that text detection can be completely freed. Specifically, we integrate IFA into the two most prevailing text recognition streams (attention-based and CTC-based) and propose attention-guided dense prediction (ADP) and Extended CTC (ExCTC). Furthermore, the Wasserstein-based Hollow Aggregation Cross-Entropy (WH-ACE) is proposed to suppress negative predictions to assist in training ADP and ExCTC. We experimentally demonstrate that IFA achieves state-of-the-art performance on end-to-end document recognition tasks while maintaining the fastest speed, and ADP and ExCTC complement each other on the perspective of different application scenarios. Code will be available at https://github.com/WangTianwei/Implicit-feature-alignment.

Xiaoxue Chen, Lianwen Jin, Yuanzhi Zhu et al.

The history of text can be traced back over thousands of years. Rich and precise semantic information carried by text is important in a wide range of vision-based application scenarios. Therefore, text recognition in natural scenes has been an active research field in computer vision and pattern recognition. In recent years, with the rise and development of deep learning, numerous methods have shown promising in terms of innovation, practicality, and efficiency. This paper aims to (1) summarize the fundamental problems and the state-of-the-art associated with scene text recognition; (2) introduce new insights and ideas; (3) provide a comprehensive review of publicly available resources; (4) point out directions for future work. In summary, this literature review attempts to present the entire picture of the field of scene text recognition. It provides a comprehensive reference for people entering this field, and could be helpful to inspire future research. Related resources are available at our Github repository: https://github.com/HCIILAB/Scene-Text-Recognition.

Zecheng Xie, Yaoxiong Huang, Yuanzhi Zhu et al.

In this paper, we propose a novel method, aggregation cross-entropy (ACE), for sequence recognition from a brand new perspective. The ACE loss function exhibits competitive performance to CTC and the attention mechanism, with much quicker implementation (as it involves only four fundamental formulas), faster inference\back-propagation (approximately O(1) in parallel), less storage requirement (no parameter and negligible runtime memory), and convenient employment (by replacing CTC with ACE). Furthermore, the proposed ACE loss function exhibits two noteworthy properties: (1) it can be directly applied for 2D prediction by flattening the 2D prediction into 1D prediction as the input and (2) it requires only characters and their numbers in the sequence annotation for supervision, which allows it to advance beyond sequence recognition, e.g., counting problem. The code is publicly available at https://github.com/summerlvsong/Aggregation-Cross-Entropy.

Yuqing Wang, Zhijie Lin, Yao Teng et al.

Autoregressive visual generation models typically rely on tokenizers to compress images into tokens that can be predicted sequentially. A fundamental dilemma exists in token representation: discrete tokens enable straightforward modeling with standard cross-entropy loss, but suffer from information loss and tokenizer training instability; continuous tokens better preserve visual details, but require complex distribution modeling, complicating the generation pipeline. In this paper, we propose TokenBridge, which bridges this gap by maintaining the strong representation capacity of continuous tokens while preserving the modeling simplicity of discrete tokens. To achieve this, we decouple discretization from the tokenizer training process through post-training quantization that directly obtains discrete tokens from continuous representations. Specifically, we introduce a dimension-wise quantization strategy that independently discretizes each feature dimension, paired with a lightweight autoregressive prediction mechanism that efficiently model the resulting large token space. Extensive experiments show that our approach achieves reconstruction and generation quality on par with continuous methods while using standard categorical prediction. This work demonstrates that bridging discrete and continuous paradigms can effectively harness the strengths of both approaches, providing a promising direction for high-quality visual generation with simple autoregressive modeling. Project page: https://yuqingwang1029.github.io/TokenBridge.

Yuyi Zhang, Yuanzhi Zhu, Dezhi Peng et al.

Text recognition, especially for complex scripts like Chinese, faces unique challenges due to its intricate character structures and vast vocabulary. Traditional one-hot encoding methods struggle with the representation of hierarchical radicals, recognition of Out-Of-Vocabulary (OOV) characters, and on-device deployment due to their computational intensity. To address these challenges, we propose HierCode, a novel and lightweight codebook that exploits the innate hierarchical nature of Chinese characters. HierCode employs a multi-hot encoding strategy, leveraging hierarchical binary tree encoding and prototype learning to create distinctive, informative representations for each character. This approach not only facilitates zero-shot recognition of OOV characters by utilizing shared radicals and structures but also excels in line-level recognition tasks by computing similarity with visual features, a notable advantage over existing methods. Extensive experiments across diverse benchmarks, including handwritten, scene, document, web, and ancient text, have showcased HierCode's superiority for both conventional and zero-shot Chinese character or text recognition, exhibiting state-of-the-art performance with significantly fewer parameters and fast inference speed.

Yuanzhi Zhu, Hanshu Yan, Huan Yang et al.

Generative models, particularly diffusion models, have made significant success in data synthesis across various modalities, including images, videos, and 3D assets. However, current diffusion models are computationally intensive, often requiring numerous sampling steps that limit their practical application, especially in video generation. This work introduces a novel framework for diffusion distillation and distribution matching that dramatically reduces the number of inference steps while maintaining-and potentially improving-generation quality. Our approach focuses on distilling pre-trained diffusion models into a more efficient few-step generator, specifically targeting video generation. By leveraging a combination of video GAN loss and a novel 2D score distribution matching loss, we demonstrate the potential to generate high-quality video frames with substantially fewer sampling steps. To be specific, the proposed method incorporates a denoising GAN discriminator to distil from the real data and a pre-trained image diffusion model to enhance the frame quality and the prompt-following capabilities. Experimental results using AnimateDiff as the teacher model showcase the method's effectiveness, achieving superior performance in just four sampling steps compared to existing techniques.

Yuanzhi Zhu, Xi Wang, Stéphane Lathuilière et al.

One-step generators distilled from Masked Diffusion Models (MDMs) compress multiple sampling steps into a single forward pass, enabling efficient text and image synthesis. However, they suffer two key limitations: they inherit modeling bias from the teacher, and their discrete token outputs block gradient flow, preventing post-distillation refinements such as adversarial training, reward-based fine-tuning, and Test-Time Embedding Optimization (TTEO). In this work, we introduce soft embeddings, a simple relaxation that replaces discrete tokens with the expected embeddings under the generator's output distribution. Soft embeddings preserve representation fidelity for one-step discrete generator while providing a fully differentiable continuous surrogate that is compatible with teacher backbones and tokenizer decoders. Integrating soft embeddings into the Di[M]O distillation framework (denoted Soft-Di[M]O) makes one-step generators end-to-end trainable and enables straightforward application of GAN-based refinement, differentiable reward fine-tuning, and TTEO. Empirically, across multiple MDM teachers (e.g., MaskBit, MaskGen), Soft-Di[M]O achieves state-of-the-art one-step results: improved class-to-image performance, a one-step FID of 1.56 on ImageNet-256 with GAN-based refinement, along with higher GenEval and HPS scores on text-to-image with reward fine-tuning, and further gains from TTEO.

Shuai Bai, Yuxuan Cai, Ruizhe Chen et al.

We introduce Qwen3-VL, the most capable vision-language model in the Qwen series to date, achieving superior performance across a broad range of multimodal benchmarks. It natively supports interleaved contexts of up to 256K tokens, seamlessly integrating text, images, and video. The model family includes both dense (2B/4B/8B/32B) and mixture-of-experts (30B-A3B/235B-A22B) variants to accommodate diverse latency-quality trade-offs. Qwen3-VL delivers three core pillars: (i) markedly stronger pure-text understanding, surpassing comparable text-only backbones in several cases; (ii) robust long-context comprehension with a native 256K-token window for both text and interleaved multimodal inputs, enabling faithful retention, retrieval, and cross-referencing across long documents and videos; and (iii) advanced multimodal reasoning across single-image, multi-image, and video tasks, demonstrating leading performance on comprehensive evaluations such as MMMU and visual-math benchmarks (e.g., MathVista and MathVision). Architecturally, we introduce three key upgrades: (i) an enhanced interleaved-MRoPE for stronger spatial-temporal modeling across images and video; (ii) DeepStack integration, which effectively leverages multi-level ViT features to tighten vision-language alignment; and (iii) text-based time alignment for video, evolving from T-RoPE to explicit textual timestamp alignment for more precise temporal grounding. Under comparable token budgets and latency constraints, Qwen3-VL achieves superior performance in both dense and Mixture-of-Experts (MoE) architectures. We envision Qwen3-VL serving as a foundational engine for image-grounded reasoning, agentic decision-making, and multimodal code intelligence in real-world workflows.

Yuanzhi Zhu, Eleftherios Tsonis, Lucas Degeorge et al.

Diffusion and flow models achieve high generative quality but remain computationally expensive due to slow multi-step sampling. Distillation methods accelerate them by training fast student generators, yet most existing objectives lack a unified theoretical foundation. In this work, we propose Di-Bregman, a compact framework that formulates diffusion distillation as Bregman divergence-based density-ratio matching. This convex-analytic view connects several existing objectives through a common lens. Experiments on CIFAR-10 and text-to-image generation demonstrate that Di-Bregman achieves improved one-step FID over reverse-KL distillation and maintains high visual fidelity compared to the teacher model. Our results highlight Bregman density-ratio matching as a practical and theoretically-grounded route toward efficient one-step diffusion generation.

Yuanzhi Zhu, Xi Wang, Stéphane Lathuilière et al.

Masked Diffusion Models (MDMs) have emerged as a powerful generative modeling technique. Despite their remarkable results, they typically suffer from slow inference with several steps. In this paper, we propose Di$\mathtt{[M]}$O, a novel approach that distills masked diffusion models into a one-step generator. Di$\mathtt{[M]}$O addresses two key challenges: (1) the intractability of using intermediate-step information for one-step generation, which we solve through token-level distribution matching that optimizes model output logits by an 'on-policy framework' with the help of an auxiliary model; and (2) the lack of entropy in the initial distribution, which we address through a token initialization strategy that injects randomness while maintaining similarity to teacher training distribution. We show Di$\mathtt{[M]}$O's effectiveness on both class-conditional and text-conditional image generation, impressively achieving performance competitive to multi-step teacher outputs while drastically reducing inference time. To our knowledge, we are the first to successfully achieve one-step distillation of masked diffusion models and the first to apply discrete distillation to text-to-image generation, opening new paths for efficient generative modeling.

Jiawei Liu, Yuanzhi Zhu, Feiyu Gao et al.

Generating visual text in natural scene images is a challenging task with many unsolved problems. Different from generating text on artificially designed images (such as posters, covers, cartoons, etc.), the text in natural scene images needs to meet the following four key criteria: (1) Fidelity: the generated text should appear as realistic as a photograph and be completely accurate, with no errors in any of the strokes. (2) Reasonability: the text should be generated on reasonable carrier areas (such as boards, signs, walls, etc.), and the generated text content should also be relevant to the scene. (3) Utility: the generated text can facilitate to the training of natural scene OCR (Optical Character Recognition) tasks. (4) Controllability: The attribute of the text (such as font and color) should be controllable as needed. In this paper, we propose a two stage method, SceneVTG++, which simultaneously satisfies the four aspects mentioned above. SceneVTG++ consists of a Text Layout and Content Generator (TLCG) and a Controllable Local Text Diffusion (CLTD). The former utilizes the world knowledge of multi modal large language models to find reasonable text areas and recommend text content according to the nature scene background images, while the latter generates controllable multilingual text based on the diffusion model. Through extensive experiments, we respectively verified the effectiveness of TLCG and CLTD, and demonstrated the state-of-the-art text generation performance of SceneVTG++. In addition, the generated images have superior utility in OCR tasks like text detection and text recognition. Codes and datasets will be available.

Canjie Luo, Yuanzhi Zhu, Lianwen Jin et al.

Large amounts of labeled data are urgently required for the training of robust text recognizers. However, collecting handwriting data of diverse styles, along with an immense lexicon, is considerably expensive. Although data synthesis is a promising way to relieve data hunger, two key issues of handwriting synthesis, namely, style representation and content embedding, remain unsolved. To this end, we propose a novel method that can synthesize parameterized and controllable handwriting Styles for arbitrary-Length and Out-of-vocabulary text based on a Generative Adversarial Network (GAN), termed SLOGAN. Specifically, we propose a style bank to parameterize the specific handwriting styles as latent vectors, which are input to a generator as style priors to achieve the corresponding handwritten styles. The training of the style bank requires only the writer identification of the source images, rather than attribute annotations. Moreover, we embed the text content by providing an easily obtainable printed style image, so that the diversity of the content can be flexibly achieved by changing the input printed image. Finally, the generator is guided by dual discriminators to handle both the handwriting characteristics that appear as separated characters and in a series of cursive joins. Our method can synthesize words that are not included in the training vocabulary and with various new styles. Extensive experiments have shown that high-quality text images with great style diversity and rich vocabulary can be synthesized using our method, thereby enhancing the robustness of the recognizer.

Canjie Luo, Yuanzhi Zhu, Lianwen Jin et al.

Handwritten text and scene text suffer from various shapes and distorted patterns. Thus training a robust recognition model requires a large amount of data to cover diversity as much as possible. In contrast to data collection and annotation, data augmentation is a low cost way. In this paper, we propose a new method for text image augmentation. Different from traditional augmentation methods such as rotation, scaling and perspective transformation, our proposed augmentation method is designed to learn proper and efficient data augmentation which is more effective and specific for training a robust recognizer. By using a set of custom fiducial points, the proposed augmentation method is flexible and controllable. Furthermore, we bridge the gap between the isolated processes of data augmentation and network optimization by joint learning. An agent network learns from the output of the recognition network and controls the fiducial points to generate more proper training samples for the recognition network. Extensive experiments on various benchmarks, including regular scene text, irregular scene text and handwritten text, show that the proposed augmentation and the joint learning methods significantly boost the performance of the recognition networks. A general toolkit for geometric augmentation is available.

Tianwei Wang, Yuanzhi Zhu, Lianwen Jin et al.

Text recognition has attracted considerable research interests because of its various applications. The cutting-edge text recognition methods are based on attention mechanisms. However, most of attention methods usually suffer from serious alignment problem due to its recurrency alignment operation, where the alignment relies on historical decoding results. To remedy this issue, we propose a decoupled attention network (DAN), which decouples the alignment operation from using historical decoding results. DAN is an effective, flexible and robust end-to-end text recognizer, which consists of three components: 1) a feature encoder that extracts visual features from the input image; 2) a convolutional alignment module that performs the alignment operation based on visual features from the encoder; and 3) a decoupled text decoder that makes final prediction by jointly using the feature map and attention maps. Experimental results show that DAN achieves state-of-the-art performance on multiple text recognition tasks, including offline handwritten text recognition and regular/irregular scene text recognition.

Xiaoxue Chen, Tianwei Wang, Yuanzhi Zhu et al.

Scene text recognition has attracted particular research interest because it is a very challenging problem and has various applications. The most cutting-edge methods are attentional encoder-decoder frameworks that learn the alignment between the input image and output sequences. In particular, the decoder recurrently outputs predictions, using the prediction of the previous step as a guidance for every time step. In this study, we point out that the inappropriate use of previous predictions in existing attention mechanisms restricts the recognition performance and brings instability. To handle this problem, we propose a novel module, namely adaptive embedding gate(AEG). The proposed AEG focuses on introducing high-order character language models to attention mechanism by controlling the information transmission between adjacent characters. AEG is a flexible module and can be easily integrated into the state-of-the-art attentional methods. We evaluate its effectiveness as well as robustness on a number of standard benchmarks, including the IIIT$5$K, SVT, SVT-P, CUTE$80$, and ICDAR datasets. Experimental results demonstrate that AEG can significantly boost recognition performance and bring better robustness.