SHOE: Supervised Hashing with Output Embeddings

Abstract

We present a supervised binary encoding scheme for image retrieval that learns projections by taking into account similarity between classes obtained from output embeddings. Our motivation is that binary hash codes learned in this way improve both the visual quality of retrieval results and existing supervised hashing schemes. We employ a sequential greedy… (More)

9 Figures and Tables

Topics

  • Presentations referencing similar topics