Effective and efficient indexing in cross-modal hashing-based datasets

作者：

Highlights：

• Build the index structure upon the binary hash codes to avoid exhaustive search.

• Learn a prediction model to provide precise indexing and high quality candidates.

• Skip the reranking process based on the precise indexing process.

摘要

•Build the index structure upon the binary hash codes to avoid exhaustive search.•Learn a prediction model to provide precise indexing and high quality candidates.•Skip the reranking process based on the precise indexing process.

论文关键词：Binary embedding,Cross-modal retrieval,Inverted indexing,Learning to rank,Nearest neighbor search

论文评审过程：Received 22 April 2019, Revised 23 August 2019, Accepted 25 September 2019, Available online 1 October 2019, Version of Record 3 October 2019.

论文官网地址：https://doi.org/10.1016/j.image.2019.115650