Abstract
Single Image Super-Resolution (SISR) is a fundamental computer vision task that aims to reconstruct a high-resolution (HR) image from a low-resolution (LR) input. Transformer-based methods have achieved remarkable performance by modeling long-range dependencies in degraded images. However, their feature-intensive attention computation incurs high computational cost. To improve efficiency, most existing approaches partition images into fixed groups and restrict attention within each group. Such group-wise attention overlooks the inherent asymmetry in token similarities, thereby failing to enable flexible and token-adaptive attention computation. To address this limitation, we propose the Individualized Exploratory Transformer (IET), which introduces a novel Individualized Exploratory Attention (IEA) mechanism that allows each token to adaptively select its own content-aware and independent attention candidates. This token-adaptive and asymmetric design enables more precise information aggregation while maintaining computational efficiency. Extensive experiments on standard SR benchmarks demonstrate that IET achieves state-of-the-art performance under comparable computational complexity.
Abstract (translated)
单图像超分辨率（SISR）是计算机视觉中的一个基本任务，旨在从低分辨率（LR）输入中重建高分辨率（HR）图像。基于Transformer的方法通过建模退化图像中的长距离依赖性而实现了显著的性能提升。然而，它们密集的注意力计算导致了较高的计算成本。为了提高效率，现有的大多数方法将图像划分为固定的组，并限制每个组内的注意机制。这种分组式的注意力忽视了令牌相似性的内在不对称性，因而无法实现灵活且适应于各个令牌的注意力计算。 为了解决这一局限性，我们提出了个性化探索Transformer（IET），它引入了一种新的个性化探索注意力（IEA）机制，允许每个令牌根据自身情况自适应地选择其内容感知且独立的注意候选者。这种针对各令牌进行调整和不对称的设计，在保持计算效率的同时实现了更精确的信息聚合。 在标准超分辨率基准上的大量实验表明，在相似的计算复杂度下，IET达到了最先进的性能水平。
URL
https://arxiv.org/abs/2601.08341
