This article introduces a sliding window model for defocus deblurring that achieves the best performance to date with extremely low memory usage. Named Swintormer, the method utilizes a diffusion model to generate latent prior features that assist in restoring more detailed images. It also extends the sliding window strategy to specialized Transformer blocks for efficient inference. Additionally, we have further optimized Multiply-Accumulate operations (Macs). Compared to the currently top-performing GRL method, our Swintormer model drastically reduces computational complexity from 140.35 GMACs to 8.02 GMacs, while also improving the Signal-to-Noise Ratio (SNR) for defocus deblurring from 27.04 dB to 27.07 dB. This new method allows for the processing of higher resolution images on devices with limited memory, significantly expanding potential application scenarios. The article concludes with an ablation study that provides an in-depth analysis of the impact of each network module on final performance. The source code and model will be available at the following website: https://github.com/bnm6900030/swintormer.

翻译：本文提出了一种用于散焦去模糊的滑动窗口模型，该模型在极低内存占用下实现了迄今为止最佳性能。该方法命名为Swintormer，利用扩散模型生成潜在先验特征，以辅助恢复更细节的图像。同时，它将滑动窗口策略扩展到专用Transformer模块以实现高效推理。此外，我们进一步优化了乘累加操作（Macs）。与当前性能最优的GRL方法相比，我们的Swintormer模型将计算复杂度从140.35 GMACs大幅降低至8.02 GMacs，同时将散焦去模糊的信噪比（SNR）从27.04 dB提升至27.07 dB。这种新方法使得在内存有限的设备上处理更高分辨率图像成为可能，显著扩展了潜在应用场景。文章最后通过消融实验，深入分析了各网络模块对最终性能的影响。源代码和模型将可在以下网站获取：https://github.com/bnm6900030/swintormer。