Unsupervised Neural Quantization for Compressed-Domain Similarity Search
This addresses the problem of efficient similarity search in image retrieval systems, but it is incremental as it builds on existing quantization approaches with a deep learning twist.
The paper tackles unsupervised visual descriptor compression for large-scale image retrieval by introducing a deep neural network architecture based on multi-codebook quantization, which outperforms previous state-of-the-art methods by a large margin on several datasets.
We tackle the problem of unsupervised visual descriptors compression, which is a key ingredient of large-scale image retrieval systems. While the deep learning machinery has benefited literally all computer vision pipelines, the existing state-of-the-art compression methods employ shallow architectures, and we aim to close this gap by our paper. In more detail, we introduce a DNN architecture for the unsupervised compressed-domain retrieval, based on multi-codebook quantization. The proposed architecture is designed to incorporate both fast data encoding and efficient distances computation via lookup tables. We demonstrate the exceptional advantage of our scheme over existing quantization approaches on several datasets of visual descriptors via outperforming the previous state-of-the-art by a large margin.