YFCC100M-HNfc6: A Large-Scale Deep Features Benchmark for Similarity Search

被引：15

作者：

Amato, Giuseppe ^{[1
]}

Falchi, Fabrizio ^{[1
]}

Gennaro, Claudio ^{[1
]}

Rabitti, Fausto ^{[1
]}

机构：

[1] ISTI CNR, Via G Moruzzi 1, I-56124 Pisa, Italy

来源：

SIMILARITY SEARCH AND APPLICATIONS, SISAP 2016 | 2016年 / 9939卷

关键词：

Similarity search; Deep features; Content-based image retrieval; Convolutional neural networks; YFCC100M;

D O I：

10.1007/978-3-319-46759-7_15

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper, we present YFCC100M-HNfc6, a benchmark consisting of 97M deep features extracted from the Yahoo Creative Commons 100M (YFCC100M) dataset. Three type of features were extracted using a state-of-the-art Convolutional Neural Network trained on the ImageNet and Places datasets. Together with the features, we made publicly available a set of 1,000 queries and k-NN results obtained by sequential scan. We first report detailed statistical information on both the features and search results. Then, we show an example of performance evaluation, performed using this benchmark, on the MI-File approximate similarity access method.

引用

页码：196 / 209

页数：14