NSD-Imagery

NSD-Imagery数据集是NSD自然场景数据集的扩展，包含人类在进行心理图像任务时的功能性磁共振成像(fMRI)活动。该数据集为评估视觉解码模型在内部生成的视觉表示上的表现提供了基准。数据集涵盖了不同复杂程度的刺激，包括简单几何形状、自然场景和抽象视觉特征的概念。NSD-Imagery数据集的发布填补了从视觉到心理图像解码方法的泛化研究的空白，为开发实用应用提供了重要资源。The NSD-Imagery dataset, an extension of the NSD Natural Scenes Dataset, contains functional magnetic resonance imaging (fMRI) recordings of human brain activity during mental imagery tasks. This dataset serves as a benchmark for evaluating the performance of visual decoding models on internally generated visual representations. The dataset covers stimuli with varying levels of complexity, including simple geometric shapes, natural scenes, and concepts of abstract visual features. The release of the NSD-Imagery dataset fills a critical gap in generalization research for visual-to-mental imagery decoding methods, providing a valuable resource for the development of practical applications.

搜集汇总

背景与挑战

背景概述

NSD-Imagery数据集是NSD自然场景数据集的扩展，包含人类进行心理图像任务时的fMRI活动，用于评估视觉解码模型在内部生成的视觉表示上的表现。它涵盖从简单几何形状到自然场景和抽象视觉特征的刺激，填补了从视觉到心理图像解码方法泛化研究的空白。