Supervised mid-level features for word image representation

By Albert Gordo. 2014

108 citations recorded by the source corpus.

This paper addresses the problem of learning word image representations: given the cropped image of a word, we are interested in finding a descriptive, robust, and compact fixed-length representation. Machine learning techniques can then be supplied with these representations to produce models useful for word retrieval or recognition tasks. Although many works have focused on the machine learning aspect once a global representation has been produced, little work has been devoted to the construction of those base image representations: most works use standard coding and aggregation techniques directly on top of standard computer vision features such as SIFT or HOG. We propose to learn local mid-level features suitable for building word image representations. These features are learnt by leveraging character bounding box annotations on a small set of training images. However, contrary to other approaches that use character bounding box information, our approach does not rely on detecting the individual characters explicitly at testing time. Our local midlevel features can then be aggregated to produce a global word image signature. When pairing these features with the recent word attributes framework of [4], we obtain results comparable with or better than the state-of-the-art on matching and recognition tasks using global descriptors of only 96 dimensions.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Deep Structured Output Learning for Unconstrained Text Recognition (2014, methodology): “Almaz´an et al. (2014) and more recently Gordo (2014) also formluate joint embedding spaces, achieving impressive results with minimal training data.”
Recursive Recurrent Nets with Attention Modeling for OCR in the Wild (2016, background): “Some early attempts [26, 53, 10] try to learn local mid-level”
ASTER: An Attentional Scene Text Recognizer with Flexible Rectification (2019, unknown): “0 - - - - Gordo [16] 93.”
Focusing Attention: Towards Accurate Text Recognition in Natural Images (2017, unknown): “0 − − − Gordo [7] 93.”
Scene Text Detection and Recognition: The Deep Learning Era (2018, background): “Rodriguez-Serrano et al. (2013), Rodriguez-Serrano et al. (2015), Gordo (2015), Almazán et al. (2014) utilize label embedding to directly performmatching between strings and images.”
ESIR: End-To-End Scene Text Recognition via Iterative Image Rectification (2018, unknown): “Gordo [10] [-] 93.”
Symmetry-Constrained Rectification Network for Scene Text Recognition (2019, unknown): “75.9 - 88.5 80.3 - - - - - Rodr´ıguez-Serrano et al. [42] 76.1 57.4 - 70.0 - - - - - - - - Jaderberg et al. [22] - - - 86.1 - 96.2 91.5 - - - - - Su and Lu [47] - - - 83.0 - 92.0 82.0 - - - - - Gordo [12] 93.3 86.6 - 91.8 - - - - - - - - Jaderberg et al. [20] 97.1 92.7 - 95.4 80.7 98.7 98.6 93.1 90.”
An End-to-End Trainable Neural Network for Image-Based Sequence Recognition and Its Application to Scene Text Recognition (2015, background): “Yao et al. [36] and Gordo et al. [14] used mid-level features for scene text recognition.”
Robust Scene Text Recognition with Automatic Rectification (2016, unknown): “0 - Gordo [12] 93.”
Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes (2018, unknown): “0 - - - - Gordo [19] 93.”
A Multi-Object Rectified Attention Network for Scene Text Recognition (2019, unknown): “- - Yao et al. [52] 80.2 69.3 - 75.9 - 88.5 80.3 - - R.-Serrano et al. [38] 76.1 57.4 - 70.0 - - - - - Jaderberg et al. [23] - - - 86.1 - 96.2 91.5 - - Su and Lu [44] - - - 83.0 - 92.0 82.0 - - Gordo [12] 93.3 86.6 - 91.8 - - - - - Jaderberg et al. [21] 95.5 89.6 - 93.2 71.7 97.8 97.0 89.6 81.8 Jade”