R-PHOC: Segmentation-Free Word Spotting using CNN

TL;DR

This paper introduces R-PHOC, a CNN-based method for segmentation-free word spotting that embeds candidate regions into a space for efficient nearest neighbor search, improving state-of-the-art results.

Contribution

It presents a novel segmentation-free approach using CNNs and PHOC embeddings for word spotting, eliminating the need for explicit segmentation.

Findings

Outperforms current state-of-the-art on GW dataset

Operates directly on images without segmentation

Performs comparably to segmentation-based methods in some cases

Abstract

This paper proposes a region based convolutional neural network for segmentation-free word spotting. Our net- work takes as input an image and a set of word candidate bound- ing boxes and embeds all bounding boxes into an embedding space, where word spotting can be casted as a simple nearest neighbour search between the query representation and each of the candidate bounding boxes. We make use of PHOC embedding as it has previously achieved significant success in segmentation- based word spotting. Word candidates are generated using a simple procedure based on grouping connected components using some spatial constraints. Experiments show that R-PHOC which operates on images directly can improve the current state-of- the-art in the standard GW dataset and performs as good as PHOCNET in some cases designed for segmentation based word spotting.