Modern neural encoders offer unprecedented text-image retrieval (TIR) accuracy. However, their high computational cost impedes an adoption to large-scale image searches. We propose a novel image ranking algorithm that uses a cascade of increasingly powerful neural encoders to progressively filter images by how well they match a given text. Our algorithm reduces lifetime TIR costs by over 3x.

翻译：现代神经编码器在文本-图像检索（TIR）中提供了前所未有的准确性。然而，其高昂的计算成本阻碍了大规模图像搜索的应用。我们提出了一种新颖的图像排序算法，该算法使用一系列能力递增的神经编码器级联，根据图像与给定文本的匹配程度逐步过滤图像。我们的算法将文本-图像检索的终身成本降低了3倍以上。