ICDAR 2013 Competition on Writer Identification

Louloudis, Georgios; Gatos, Basilis; Stamatopoulos, Nikolaos; Papandreou, A.

doi:10.1109/icdar.2013.282

Cited by 60 publications

(40 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The proposed method is evaluated on two public datasets: IAM [18] and ICDAR 2013 writer identification contest dataset (called ICDAR2013 dataset) [3]. The IAM dataset [18] contains 1539 English handwriting document images from 657 writers.…”

Section: Methodsmentioning

confidence: 99%

“…The ICDAR2013 dataset [3] consists of a training dataset and a test dataset, containing 100 writers and 250 writers, respectively. Each writer is asked to copy four pages of text in two languages (two in English and two in Greek).…”

Section: Methodsmentioning

confidence: 99%

“…The "leave-one-out" strategy means that for each handwriting document sample, the distances to all other samples of these datasets are computed [3,10]. The TOP-N performance criterion is defined as that a correct hit is considered when at least one document image of the same writer is included in the N most similar document images with K-nearest-neighbors as [3,10].…”

Section: Methodsmentioning

confidence: 99%

“…Recently, due to its importance for forensic analysis and documents authorization, text-independent writer identification has attracted an increasing interest. And a series of international writer identification contests [1][2][3] have been successfully organized. A number of various approaches for text-independent writer identification have been proposed in the previous researches.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Text-Independent Writer Identification Using Improved Structural Features

Tang

2014

Biometric Recognition

View full text Add to dashboard Cite

Abstract. This paper presents a method based on two structural features for text-independent writer identification, i.e. SIFT descriptor (SD) and triangular descriptor (TD). For SD, we modify the original SIFT algorithm to make the SD possess orientation information, called modified SIFT descriptor (MSD). Acodebookis constructed by clustering the MSDs extracted from training samples. Then the bag of word technique is used to compute a MSD histogram (MSDH) as a feature vector for writer identification. For TD, it is designed to represent the unique relationship between three selected points. A TD histogram (TDH) of the TD occurrences is computed as another feature vector by tracking the contour points of a handwriting image. The distances between MSDHs and TDHs are computed and combined as the final dissimilarity measurement for the handwriting images. Experimental results on two public challenging datasets demonstrate the efficiency of the proposed method.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Text-Independent Writer Identification Using Improved Structural Features

Tang

2014

Biometric Recognition

View full text Add to dashboard Cite

show abstract

“…4 shows handwriting samples from the CERUG-EN data set. We also evaluate the proposed method on the benchmarking dataset used on ICDAR2013 Competition on Writer Identification [23]. This data set contains 1000 handwritings written by 250 writers and each writer contributed four pages (two Greek and two English).…”

Section: A Data Setmentioning

confidence: 99%

General Pattern Run-Length Transform for Writer Identification

Schomaker

2016

2016 12th IAPR Workshop on Document Analysis Systems (DAS)

View full text Add to dashboard Cite

In this paper we present a novel textural-based feature for writer identification: the General Pattern RunLength Transform (GPRLT), which is the histogram of the runlength of any complex patterns. The GPRLT can be computed on the binary images (GPRLT bin ) or on the gray scale images (GPRLT gray ) without using any binarization or segmentation methods. Experimental results show that the GPRLT gray achieves even higher performance than the GPRLT bin for writer identification. The writer identification performance on the challenging CERUG-EN data set demonstrates that the proposed methods outperform state-of-the-art algorithms. Our source code and data set are available on www.ai.rug.nl/ ∼ sheng/dflib.

show abstract

Non-deterministic Behavior of Ranking-Based Metrics When Evaluating Embeddings

Nicolaou

Dey

Christlein

et al. 2019

Reproducible Research in Pattern Recognition

View full text Add to dashboard Cite

Embedding data into vector spaces is a very popular strategy of pattern recognition methods. When distances between embeddings are quantized, performance metrics become ambiguous. In this paper, we present an analysis of the ambiguity quantized distances introduce and provide bounds on the effect. We demonstrate that it can have a measurable effect in empirical data in state-of-the-art systems. We also approach the phenomenon from a computer security perspective and demonstrate how someone being evaluated by a third party can exploit this ambiguity and greatly outperform a random predictor without even access to the input data. We also suggest a simple solution making the performance metrics, which rely on ranking, totally deterministic and impervious to such exploits.

show abstract

ICDAR 2013 Competition on Writer Identification

Cited by 60 publications

References 14 publications

Text-Independent Writer Identification Using Improved Structural Features

Text-Independent Writer Identification Using Improved Structural Features

General Pattern Run-Length Transform for Writer Identification

Non-deterministic Behavior of Ranking-Based Metrics When Evaluating Embeddings

Contact Info

Product

Resources

About