Sketch-QNet: A Quadruplet ConvNet for Color Sketch-based Image Retrieval
It addresses a critical limitation in real-world search applications where existing methods fail to distinguish items with similar types but different attributes like color or texture.
The paper tackles the problem of discriminating weakly relevant items in similarity search by proposing a quadruplet-based architecture, specifically Sketch-QNet for color sketch-based image retrieval, achieving new state-of-the-art results.
Architectures based on siamese networks with triplet loss have shown outstanding performance on the image-based similarity search problem. This approach attempts to discriminate between positive (relevant) and negative (irrelevant) items. However, it undergoes a critical weakness. Given a query, it cannot discriminate weakly relevant items, for instance, items of the same type but different color or texture as the given query, which could be a serious limitation for many real-world search applications. Therefore, in this work, we present a quadruplet-based architecture that overcomes the aforementioned weakness. Moreover, we present an instance of this quadruplet network, which we call Sketch-QNet, to deal with the color sketch-based image retrieval (CSBIR) problem, achieving new state-of-the-art results.