O3 (Odd-One-Out Dataset)

一组“在野外”聚集的逼真的奇数刺激。 Odd-One-Out (O3) 数据集中的每张图像都描绘了一个场景，该场景具有多个外观相似的对象（干扰物）和一个或多个特征维度（例如颜色、形状、大小）不同的单例（目标）。调整所有图像的大小，以使较大的尺寸为 1024 像素。目标代表大约。 400种常见的物体类型，如鲜花、糖果、鸡蛋、树叶、瓷砖和鸟类。为目标和干扰物提供了逐像素蒙版。使用 CVAT 生成注释。

A set of realistic in-the-wild odd-one-out stimuli. Each image in the Odd-One-Out (O3) dataset depicts a scene containing multiple visually similar objects (distractors) and a single singleton (target) that differs in one or more feature dimensions (e.g., color, shape, size). All images are resized such that their longer side is 1024 pixels. The targets represent approximately 400 common object categories, such as flowers, candies, eggs, leaves, tiles, and birds. Pixel-wise masks are provided for both the targets and distractors. Annotations are generated using CVAT.