Sub-Selective Quantization for Large-Scale Image Search

Authors: Yeqing Li, Chen Chen, Wei Liu, Junzhou Huang

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments are carried out on three image benchmarks with up to one million samples, corroborating the efficacy of the sub-selective quantization method in terms of image retrieval.
Researcher Affiliation Collaboration 1 University of Texas at Arlington, Arlington, TX, 76019, USA 2IBM T. J. Watson Research Center, 10598, NY, USA
Pseudocode Yes Algorithm 1 PCA Quantization (PCAQ) ... Algorithm 2 Iterative Quantization (ITQ) ... Algorithm 3 ITQ with Sub-Selection (ITQ-SS)
Open Source Code No The paper does not provide concrete access information (e.g., specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described in this paper.
Open Datasets Yes We evaluate the Sub-selective Quantization approaches on three public datasets: CIFAR (Krizhevsky and Hinton 2009) 1, MNIST2 and Tiny-1M (Wang, Kumar, and Chang 2012). 1http://www.cs.toronto.edu/ kriz/cifar.html 2http://yann.lecun.com/exdb/mnist/
Dataset Splits No The paper specifies training and test splits for the datasets but does not explicitly mention or provide details for a validation split.
Hardware Specification Yes All our experiments were conducted on a desktop computer with a 3.4GHz Intel Core i7 and 12GB RAM.
Software Dependencies No The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment.
Experiment Setup No The paper mentions the sub-selection ratio and method initialization but does not provide other concrete experimental setup details, such as the specific number of iterations (N or p) used for ITQ/ITQ-SS in the experiments, which is a key hyperparameter.