AsConvSR：基于组合卷积的快速轻量超分辨率网络
AsConvSR: Fast and Lightweight Super-Resolution Network with Assembled Convolutions
AI总结：
针对高分辨率图像带宽受限及超分网络实时性难题，提出基于组合卷积的快速轻量超分网络AsConvSR，性能优于现有高效超分模型，获NTIRE 2023实时超分赛道1冠军。
摘要：
近年来，720p（高清）、1080p（全高清）和4K（超高清）分辨率的视频与图像在电视、手机、VR等显示设备上愈发普及。但受互联网带宽限制，这类高分辨率图像无法达到预期视觉效果，也给超分辨率网络实现实时性能带来巨大挑战。针对这一难题，我们探索了像素反洗牌、重复上采样、移除局部跳跃连接等多种高效网络设计方案，提出了一种快速轻量的超分辨率网络。此外，通过分析分治思想在超分辨率中的应用，我们提出了组合卷积（assembled convolutions），可根据输入特征自适应调整卷积核。实验表明，我们的方法优于所有现有最先进的高效超分辨率模型，在运行速度和成像质量上均取得最优结果。同时，该方法还在NTIRE 2023实时超分辨率赛道1（×2倍）中夺得第一名。代码将发布于https://gitee.com/mindspore/models/tree/master/research/cv/AsConvSR。
英文摘要：
In recent years, videos and images in 720p (HD), 1080p (FHD) and 4K (UHD) resolution have become more popular for display devices such as TVs, mobile phones and VR. However, these high resolution images cannot achieve the expected visual effect due to the limitation of the internet bandwidth, and bring a great challenge for super-resolution networks to achieve real-time performance. Following this challenge, we explore multiple efficient network designs, such as pixel-unshuffle, repeat upscaling, and local skip connection removal, and propose a fast and lightweight super-resolution network. Furthermore, by analyzing the applications of the idea of divide-and-conquer in super-resolution, we propose assembled convolutions which can adapt convolution kernels according to the input features. Experiments suggest that our method outperforms all the state-of-the-art efficient super-resolution models, and achieves optimal results in terms of runtime and quality. In addition, our method also wins the first place in NTIRE 2023 Real-Time Super-Resolution - Track 1 ($\times$2). The code will be available at https://gitee.com/mindspore/models/tree/master/research/cv/AsConvSR
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
