Search

An autoregressive (AR) decoder for scene text recognition (STR) requires numerous generation steps to decode a text image character by character but can yield high recognition accuracy. On the other hand, a non-autoregressive (NAR) decoder generates all characters in a single generation but suffers from a loss of recognition accuracy. This is because, unlike the former, the latter assumes that the predicted characters are conditionally independent. This paper presents a Partially Autoregressive Scene Text Recognition (PARSTR) method that unifies both AR and NAR decoding within the same model. To reduce decoding steps while maintaining recognition accuracy, we devise two decoding strategies: b-first and b-ahead, reducing the decoding steps to approximately band by a factor of b, respectively. The experimental results demonstrate that our PARSTR models using the devised decoding strategies present a balanced compromise between efficiency and recognition accuracy compared to the fully AR and NAR decoding approaches. Specifically, the experimental results on public benchmark STR datasets demonstrate the potential to reduce decoding steps down to at most five steps and by a factor of five under the b-first and b-ahead decoding schemes, respectively, while having a slight reduction of total word recognition accuracy of less than or equal to 0.5%.

The integration of an attention mechanism has played a crucial role in many recent scene text recognition (STR) methods. It enables the capture of spatial feature dependencies (known as self-attention) and the identification of relevant features while predicting a character (known as cross-attention). However, computations and memory requirements in the self-attention and cross-attention layers increase quadratically and linearly with the feature map size, respectively, leading to a computational bottleneck in low-resource environments. But, is it necessary to attend to the entire feature maps? On the other hand, text in a natural scene is continuous and oriented in a specific direction, and it does not occupy the entire image. Therefore, utilizing only a small salient subset of features in text regions is sufficient for accurately predicting characters. Based on this salient feature selection, we propose a reduced-complexity scene text recognition framework that significantly reduces model complexities and memory requirements in the self-attention and cross-attention layers. We validate the proposed framework by employing a convolutional STR architecture with both connectionist temporal classification and transformer decoders. Through the model complexity and performance analyses on public benchmark datasets, we demonstrate that the proposed method can substantially reduce model complexities while still maintaining reasonably robust recognition accuracy.

In the future, it is possible that universities could use sensing devices to help manage students and classroom disruptions during pandemics. Wearable sensing devices have shown the capability to identify oncoming illnesses. In particular, the Oura Ring, a smart-ring with sleep-tracking features, can detect cases of COVID-19 before users are aware of it. While these devices show promise, they are only effective for pandemic preparedness and response if users pay attention to the outputs of the devices and consistently use them. In this article, we discuss the results of a year-long in-the-wild study with 35 participants at a university who wore Oura Rings. After an orientation, the participants wore their rings with no restrictions or minimum wearing requirements. By retroactively looking at how participants used the rings for monitoring their health, we identify strategies and potential problems with employing wearables for health monitoring in universities.

This paper proposes a fast recognition method based on generalized similarity measure (GSM). The GSM achieves good recognition accuracy for face recognition, but has a scalability problem. Because the GSM method requires the similarity measures between a query and all samples to be calculated, the computational cost for recognition is in proportion to the number of samples. A reasonable approach to avoiding calculating all the similarity measures is to limit the number of samples used for calculation. Although approximate nearest neighbor search (ANNS) methods take this approach, they cannot be applied to the GSM-based method directly because they assume that similarity measure is the Euclidean distance. The proposed method embeds the GSM into the Euclidean distance so that it may be applied in existing ANNS methods. We conducted experiments on face, object, and character datasets, and the results show that the proposed method achieved fast recognition without dropping the accuracy.

The eye tracking technology is used for four decades for studying reading behavior. The applications are various: estimating the reader comprehension, identifying the reader, summarizing a read document, creating a reading-life log, etc. The gaze data used in such applications has to be accurate enough to perform the analysis. In order to improve the accuracy, most of the experiments are set up with restrictive conditions such as using a head fixation and a professional eye tracker. It implies that the results are valid only in restrictive laboratory settings and an unrealistic small error is produced by the experiment. However, the use of affordable eye trackers in realistic conditions of reading leads to large errors in the recordings. We propose a new algorithm to correct the vertical error and to align the gazes with the text. The proposed algorithm is robust to rereading and skipping some parts of text, contrary to all the other algorithms of the state of the art. We show that up to 69 % of the gazes are aligned with the correct text lines.

In an increasingly interconnected world with more and more technology surrounding us, more and more people end up with psychological and mental health issues. Mobile Computing technologies is often introduced into society without knowing about impact and consequences. We need new mobile tools for everyday life to better manage our attention and give us introspection into our cognitive self. There are patterns in human physiological signals (facial expressions, heart rate, nasal temperature, eye movements, blinking, etc.) that can reveal information about cognitive functions of individuals and groups.

This document is a collection of four working group reports in the areas of digital libraries, document image retrieval, layout analysis, and Web document analysis. These reports were the outcome of discussions by participants at the Fifth IAPR International Workshop on Document Analysis Systems held in Princeton, NJ on 19-21 August 2002.This document is a collection of four working group reports in the areas of digital libraries, document image retrieval, layout analysis, and Web document analysis. These reports were the outcome of discussions by participants at the Fifth IAPR International Workshop on Document Analysis Systems held in Princeton, NJ on 19-21 August 2002.

This paper presents a method of page segmentation based on the approximated area Voronoi diagram. The characteristics of the proposed method are as follows: (1) The Voronoi diagram enables us to obtain the candidates of boundaries of document components from page images with non-Manhattan layout and a skew. (2) The candidates are utilized to estimate the intercharacter and interline gaps without the use of domain-specific parameters to select the boundaries. From the experimental results for 128 images with non-Manhattan layout and the skew of 0°∼45° as well as 98 images with Manhattan layout, we have confirmed that the method is effective for extraction of body text regions, and it is as efficient as other methods based on connected component analysis.