# Image Retrieval and Pattern Spotting using Siamese Neural Network

## TL;DR

This paper introduces a Siamese Neural Network-based method for image retrieval and pattern spotting in document collections, achieving high accuracy without manual feature engineering.

## Contribution

The paper proposes a novel similarity-based representation learned via Siamese Neural Networks for document image retrieval and pattern spotting.

## Key findings

- Achieves 0.94 mAP for retrieval
- Achieves 0.83 mAP for pattern spotting
- Demonstrates robustness across different feature map sizes

## Abstract

This paper presents a novel approach for image retrieval and pattern spotting in document image collections. The manual feature engineering is avoided by learning a similarity-based representation using a Siamese Neural Network trained on a previously prepared subset of image pairs from the ImageNet dataset. The learned representation is used to provide the similarity-based feature maps used to find relevant image candidates in the data collection given an image query. A robust experimental protocol based on the public Tobacco800 document image collection shows that the proposed method compares favorably against state-of-the-art document image retrieval methods, reaching 0.94 and 0.83 of mean average precision (mAP) for retrieval and pattern spotting (IoU=0.7), respectively. Besides, we have evaluated the proposed method considering feature maps of different sizes, showing the impact of reducing the number of features in the retrieval performance and time-consuming.

---
Source: https://tomesphere.com/paper/1906.09513