End-to-End Interpretation of the French Street Name Signs Dataset

TL;DR

This paper introduces a large, complex dataset of French street name signs and demonstrates an end-to-end deep learning approach for extracting street names from Google Street View images, enabling improved map labeling.

Contribution

The paper presents the FSNS dataset with over a million images and an end-to-end deep learning model for street name recognition, advancing automated map annotation.

Findings

Deep network trained on FSNS achieves high accuracy.

End-to-end approach simplifies street name extraction pipeline.

Dataset enables future research in street sign recognition.

Abstract

We introduce the French Street Name Signs (FSNS) Dataset consisting of more than a million images of street name signs cropped from Google Street View images of France. Each image contains several views of the same street name sign. Every image has normalized, title case folded ground-truth text as it would appear on a map. We believe that the FSNS dataset is large and complex enough to train a deep network of significant complexity to solve the street name extraction problem "end-to-end" or to explore the design trade-offs between a single complex engineered network and multiple sub-networks designed and trained to solve sub-problems. We present such an "end-to-end" network/graph for Tensor Flow and its results on the FSNS dataset.