Year: 2020
Type: preprint
Abstract: Scene text recognition models have advanced greatly in recent years. Inspired by human reading we characterize two important scene text recognition models by measuring their domains i.e. the range of stimulus images that they can read. The domain specifies the ability of readers to generalize to different word lengths, fonts, and amounts of occlusion. These metrics identify strengths and weaknesses of existing models. Relative to the attention-based (Attn) model, we discover that the connectionist temporal classification (CTC) model is more robust to noise and occlusion, and better at generalizing to different word lengths. Further, we show that in both models, adding noise to training images yields better generalization to occlusion. These results demonstrate the value of testing models till they break, complementing the traditional data science focus on optimizing performance.
Source: arXiv (Cornell University)
Authors: Siddiqui, Sahar, Sizikova, Elena, Roig, Gemma, Majaj, Najib J., Pelli, Denis G.
Institutions: New York University, Goethe University Frankfurt
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3040254674