Multimodal interactive transcription of text images

Toselli, Alejandro H.; Romero, Verónica; Pastor, María Luisa Carrió; Vidal, Enrique

doi:10.1016/j.patcog.2009.11.019

Cited by 77 publications

(60 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The results in the lower part of the table are the best results reported so far for IAM. España-Boquera [23] used neural networks to perform particular preprocessing [24] developed an HMM-based system using gray-scale and gradient features. Graves [25] used an LSTM recurrent neural network with a CTC output layer.…”

Section: Resultsmentioning

confidence: 99%

Open vocabulary handwriting recognition using combined word-level and character-level language models

Kozielski

Rybach

Hahn

et al. 2013

2013 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

In this paper, we present a unified search strategy for open vocabulary handwriting recognition using weighted finite state transducers. Additionally to a standard word-level language model we introduce a separate n-gram character-level language model for out-of-vocabulary word detection and recognition. The probabilities assigned by those two models are combined into one Bayes decision rule. We evaluate the proposed method on the IAM database of English handwriting. An improvement from 22.2% word error rate to 17.3% is achieved comparing to the closed-vocabulary scenario and the best published result.

show abstract

Section: Resultsmentioning

confidence: 99%

Open vocabulary handwriting recognition using combined word-level and character-level language models

Kozielski

Rybach

Hahn

et al. 2013

2013 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

show abstract

“…In [17], a multimodal interactive scenario where the user and the system collaborate to generate a better solution was presented for handwritten text recognition (HTR). The situation in layout analysis and text line segmentation is rather similar.…”

Section: Related Workmentioning

confidence: 99%

Escritoire: A Multi-touch Desk with e-Pen Input for Capture, Management and Multimodal Interactive Transcription of Handwritten Documents

Martín-Albo¹,

Romero²,

Vidal³

2015

Pattern Recognition and Image Analysis

Self Cite

View full text Add to dashboard Cite

Abstract. A large quantity of documents used every day are still handwritten. However, it is interesting to transform each of these documents into its digital version for managing, archiving and sharing. Here we present Escritoire, a multitouch desk that allows the user to capture, transcribe and work with handwritten documents. The desktop is continuously monitored using two cameras. Whenever the user makes a specific hand gesture over a paper, Escritoire proceeds to take an image. Then, the capture is automatically preprocesses, obtaining as a result an improved representation. Finally, the text image is transcribed using automatic techniques and finally the transcription is displayed on Escritoire.

show abstract

“…The process starts when the HTR system proposes a full transcription of a feature vector sequence x, extracted from a handwritten text line image. The user validates an initial part of this transcription, p , which is error-free and introduces a correct word, v, thereby producing correct transcription prefix, p = p v. Then, the HTR system takes into account the available information to suggest a new suitable continuation suffix, s. This process is repeated until a full correct transcription of x is accepted by the user [7].…”

Section: Catti Overviewmentioning

confidence: 99%

“…In previous works [7,5], a more effective, interactive on-line approach was presented. This approach, called "Computer Assisted Transcription of Handwritten Text Images" (CATTI), combines the accuracy ensured by the human transcriber with the efficiency of the HTR systems to obtain final perfect transcriptions.…”

Section: Introductionmentioning

confidence: 99%

“…So far, human corrective feedback for CATTI has been studied at two different levels: a) whole-word interactions (both typed and handwritten using an e-pen interface [7]) and b) (typed) character-level corrections [5]. According to the results of these works, keystroke corrections can save a significant quantity of human effort with respect to whole-word corrections, while multimodal, e-pen interaction seems more ergonomic for human transcribers, which is a key point in the design of friendly and usable user interfaces.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Character-Level Interaction in Multimodal Computer-Assisted Transcription of Text Images

Martín-Albo

Romero

Toselli

et al. 2011

Pattern Recognition and Image Analysis

Self Cite

View full text Add to dashboard Cite

Abstract. To date, automatic handwriting text recognition systems are far from being perfect and heavy human intervention is often required to check and correct the results of such systems. As an alternative, an interactive framework that integrates the human knowledge into the transcription process has been presented in previous works. In this work, multimodal interaction at character-level is studied. Until now, multimodal interaction had been studied only at whole-word level. However, character-level pen-stroke interactions may lead to more ergonomic and friendly interfaces. Empirical tests show that this approach can save significant amounts of user effort with respect to both fully manual transcription and non-interactive post-editing correction.

show abstract

Multimodal interactive transcription of text images

Cited by 77 publications

References 13 publications

Open vocabulary handwriting recognition using combined word-level and character-level language models

Open vocabulary handwriting recognition using combined word-level and character-level language models

Escritoire: A Multi-touch Desk with e-Pen Input for Capture, Management and Multimodal Interactive Transcription of Handwritten Documents

Character-Level Interaction in Multimodal Computer-Assisted Transcription of Text Images

Contact Info

Product

Resources

About