Parametric and nonparametric residual vector quantization optimizations for ANN search

Compact Hash Codes for Efficient Visual Descriptors Retrieval in Large Scale Databases

“…6) and GIST1M (Fig. 7) datasets against several methods presented in Section II: Product Quantization (ADC and IVFADC) [1], PQ-RO [19], PQ-RR [19], Cartesian k-means [15], OPQ-P [16], [51], OPQ-NP [16], [51], LOPQ [18], a non-exhaustive adaptation of OPQ [16], called I-OPQ [18], RVQ [52] , RVQ-P [19] and RVQ-NP [19].…”

“…The length of the final code is given by m × log 2 k * . PQ-RO [19] is the Product Quantization approach with data projection by randomly order dimensions.…”

“…PQ-RR [19] is the Product Quantization approach with data projection by both PCA and randomly rotation.…”

“…Kalantidis and Avrithis [18] have presented a simple vector quantizer (LOPQ) which uses a local optimization over a rotation and a space decomposition and apply a parametric solution that assumes a normal distribution. More recently, Guo et al [19] have improved over OPQ and LOPQ adding two quantization distortion properties of the Residual Vector Quantization (RVQ) model, that tries to restore quantization distortion errors instead of reducing it.…”

“…Recall@R on SIFT1M. Comparison between our method (m-k-meanst 1 , m-k-means-n 1 with n = 32, m-k-means-t 2 and m-k-means-n 2 with n = 32), the Product Quantization method (PQ ADC and PQ IVFADC) [1], Cartesian k-means method (ck-means) [15], a non-exhaustive adaptation of the optimized product quantization method (I-OPQ), a locally optimized product quantization method (LOPQ) [18], OPQ-P, and OPQ-NP [16], [51], and PQ-RO, PQ-RR, RVQ-P, and RVQ-NP [19]. Fig.…”