Queries Processing
Last updated
Was this helpful?
Last updated
Was this helpful?
The inverted file from , Sivic & Zisserman, ICCV 2003. This is the key to non-exhaustive search in large datasets. Otherwise all searches would need to scan all elements in the index, which is prohibitive even if the operation to apply for each element is fast
The product quantization (PQ) method from , JĂŠgou & al., PAMI 2011. This can be seen as a lossy compression technique for high-dimensional vectors, that allows relatively accurate reconstructions and distance computations in the compressed domain.
The three-level quantization (IVFADC-R aka IndexIVFPQR) method from , Tavenard & al., ICASSP'11.
The inverted multi-index from , Babenko & Lempitsky, CVPR 2012. This method greatly improves the speed of inverted indexing for fast/less accurate operating points.
The optimized PQ from , He & al, CVPR 2013. This method can be seen as a linear transformation of the vector space to make it more amenable for indexing with a product quantizer.
The pre-filtering of product quantizer distances from , Douze & al., ECCV 2016. This technique performs a binary filtering stage before computing PQ distances.
The GPU implementation and fast k-selection is described in , Johnson & al, ArXiv 1702.08734, 2017
The HNSW indexing method from , Malkov & al., ArXiv 1603.09320, 2016
The in-register vector comparisons from , AndrĂŠ et al, PAMI'19, also used in , Guo, Sun et al, ICML'20.
A general paper about product quantization and related methods: , Yusuke Matsui, Yusuke Uchida, HervĂŠ JĂŠgou, Shinâichi Satoh, ITE transactions on MTA, 2018.