Exploring Architectures for CNN-Based Word Spotting
This work addresses the problem of optimizing CNN architectures for word spotting in document analysis, providing insights for researchers and practitioners, though it is incremental as it compares existing methods.
The paper investigates how CNN architecture complexity affects word spotting performance, finding that complex models like ResNet and DenseNet improve results on challenging datasets like IAM Offline Database but offer no advantage on easier ones like George Washington Database.
The goal in word spotting is to retrieve parts of document images which are relevant with respect to a certain user-defined query. The recent past has seen attribute-based Convolutional Neural Networks take over this field of research. As is common for other fields of computer vision, the CNNs used for this task are already considerably deep. The question that arises, however, is: How complex does a CNN have to be for word spotting? Are increasingly deeper models giving increasingly better results or does performance behave asymptotically for these architectures? On the other hand, can similar results be obtained with a much smaller CNN? The goal of this paper is to give an answer to these questions. Therefore, the recently successful TPP-PHOCNet will be compared to a Residual Network, a Densely Connected Convolutional Network and a LeNet architecture empirically. As will be seen in the evaluation, a complex model can be beneficial for word spotting on harder tasks such as the IAM Offline Database but gives no advantage for easier benchmarks such as the George Washington Database.