Abstract:
Feature extraction plays a pivotal role in the context of single image super-resolution. Nonetheless, relying on a single feature extraction method often undermines the full potential of feature representation, hampering the model\'s overall performance. To tackle this issue, this study introduces the wide-activation feature distillation network (WFDN), which realizes single image super-resolution through dual-path learning. Initially, a dual-path parallel network structure is employed, utilizing a residual network as the backbone and incorporating global residual connections to enhance feature exploitation and expedite network convergence. Subsequently, a feature distillation block is adopted, characterized by fast training speed and a low parameter count. Simultaneously, a wide-activation mechanism is integrated to further enhance the representational capacity of high-frequency features. Lastly, a gated fusion mechanism is introduced to weight the fusion of feature information extracted from the dual branches. This mechanism enhances reconstruction performance while mitigating information redundancy. Extensive experiments demonstrate that the proposed algorithm achieves stable and superior results compared to the state-of-the-art methods, as evidenced by quantitative evaluation metrics tests conducted on four benchmark datasets. Furthermore, our WFDN excels in reconstructing images with richer detailed textures, more realistic lines, and clearer structures, affirming its exceptional superiority and robustness.
摘要:
在单幅图像超分辨率背景下,特征提取起着举足轻重的作用。尽管如此,依靠单一的特征提取方法往往会破坏特征表示的全部潜力,妨碍模型的整体性能。为了解决这个问题,这项研究介绍了广泛的活化特征蒸馏网络(WFDN),通过双路径学习实现单幅图像的超分辨率。最初,采用双路径并行网络结构,利用剩余网络作为骨干,并结合全局剩余连接,以增强功能开发并加快网络融合。随后,采用了特征蒸馏块,其特点是训练速度快,参数计数低。同时,整合了广泛的激活机制,以进一步提高高频特征的表示能力。最后,引入门控融合机制对双分支提取的特征信息进行加权融合。该机制增强了重建性能,同时减轻了信息冗余。大量的实验表明,与最先进的方法相比,该算法获得了稳定和优越的结果,对四个基准数据集进行的定量评估指标测试证明了这一点。此外,我们的WFDN擅长重建具有更丰富详细纹理的图像,更现实的线条,更清晰的结构,肯定了其非凡的优越性和稳健性。