提出统一框架，联合生成图像与文字实现高效高保真文本图像超分辨率。
Coupled Continuous-Discrete Generation for Scene Text Image Super-Resolution
- 将图像与文本生成耦合为连续-离散联合建模，共享多模态主干网络。
- 在CTR-TSR和RealCE上均达最优性能，X4下准确率提升12.78个百分点。
- 无需外部OCR，参数量减少83%，推理速度提升99%以上，适合实际部署。
场景文本图像超分辨率（STISR）旨在从退化输入中恢复视觉上合理的外观并保留字符语义。现有方法常依赖外部先验或分离的图像与文本模型，导致误差传播和高成本多阶段推理。我们提出DualTSR，一个统一框架，将STISR建模为耦合的连续-离散生成过程：条件流匹配恢复连续图像隐变量，吸收态离散扩散重建文本令牌。两者共享多模态Transformer主干，使图像与文本状态在生成过程中持续交互，无需推理时使用外部OCR。在CTR-TSR数据集上，DualTSR在X2和X4缩放下均达到最优的FID、LPIPS、ACC和NED；在对齐的RealCE子集上，也取得最佳FID、ACC和NED，LPIPS表现接近最优。相比DiffTSR在X4下，DualTSR将ACC提升12.78个百分点，参数量从1.23B降至203M，端到端延迟从13.3秒降至132毫秒。结果表明DualTSR是准确且高效的STISR方法。
原文摘要 · Abstract (English)
Scene text image super-resolution (STISR) aims to recover visually plausible appearance while preserving character semantics from degraded inputs. Existing STISR systems often rely on externally generated priors or separate image and text models, resulting in error propagation and costly multi-stage inference. We present DualTSR, a unified framework that formulates STISR as coupled continuous-discrete generation. Conditional flow matching restores continuous image latents, while absorbing-state discrete diffusion reconstructs text tokens. Both processes share a multimodal transformer backbone, allowing the evolving image and text states to interact throughout generation without an external OCR prior at inference. On CTR-TSR, DualTSR achieves the best FID, LPIPS, ACC, and NED among the compared methods at both X2 and X4. On an aligned RealCE subset, it obtains the best FID, ACC, and NED with competitive LPIPS. Compared with DiffTSR at X4, DualTSR improves ACC by 12.78 percentage points while reducing the parameter count from 1.23B to 203M and end-to-end latency from 13.3s to 132ms. These results establish DualTSR as an accurate and efficient method for STISR.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。