Call for participation- ICPR 2018 Fraud Detection Contest
12 views
Skip to first unread message
Nicolas Sidère
Feb 15, 2018, 11:44:17 AM2/15/18
to Computational Forensics
Call for Participation
ICPR 2018 Contest on Fraud Detection
We are pleased to announce the Fraud Detection Contest in conjunction with ICPR 2018.
This contest sits at the crossroad between text and image analysis, allowing both text- and image-based approaches to discover fraud from digitally-born and digitised documents.
Since last decades, the explosion of the volume of digital documents images and development of easy-to-use consumer tools to edit these images has led to a huge increase in the number of corrupted documents. The development of many tools and methods to detect modifications has also increased but their evaluation remains a challenge.
Forensics research is a sensitive topic. Datasets are often private or unlabeled and most related works are evaluated based on undisclosed datasets. This restriction has two major consequences: results cannot be reproduced and no benchmarking can be done between approaches.
This contest was conceived in order to address these drawbacks:
● First, a public images and texts dataset of french shopping receipts containing no private information was designed so that it can be disclosed and used with no constraints.
● Second, the contest should finally presents a benchmark on major works developed since many years, both for images and texts processing methods.
The contest is open to all researchers and engineers working on detection of forgery within documents, images or texts, coming from different domains of activities (forensics, document analysis, data mining…).
ICPR 2018 Contest on Fraud Detection
We are pleased to announce the Fraud Detection Contest in conjunction with ICPR 2018.
This contest sits at the crossroad between text and image analysis, allowing both text- and image-based approaches to discover fraud from digitally-born and digitised documents.
Since last decades, the explosion of the volume of digital documents images and development of easy-to-use consumer tools to edit these images has led to a huge increase in the number of corrupted documents. The development of many tools and methods to detect modifications has also increased but their evaluation remains a challenge.
Forensics research is a sensitive topic. Datasets are often private or unlabeled and most related works are evaluated based on undisclosed datasets. This restriction has two major consequences: results cannot be reproduced and no benchmarking can be done between approaches.
This contest was conceived in order to address these drawbacks:
● First, a public images and texts dataset of french shopping receipts containing no private information was designed so that it can be disclosed and used with no constraints.
● Second, the contest should finally presents a benchmark on major works developed since many years, both for images and texts processing methods.
The contest is open to all researchers and engineers working on detection of forgery within documents, images or texts, coming from different domains of activities (forensics, document analysis, data mining…).
Tasks
Task 1: The detection of modified documents. The provided dataset will contain both genuine and modified documents. The first task aims at discovering which ones are forged.
Task 2: The spotting of this or these falsification(s) in the document (many falsifications can occur in one document), with methods applied to the image or text, or both.
Task 2: The spotting of this or these falsification(s) in the document (many falsifications can occur in one document), with methods applied to the image or text, or both.
Important dates
Registration deadline: March 30, 2018
Training dataset is available: early February, 2018
Evaluation dataset is available: April 5, 2018
Result submission: April 10, 2018
Submission of the description of methods: April 15, 2018
Conference: August 20-24, 2018
Training dataset is available: early February, 2018
Evaluation dataset is available: April 5, 2018
Result submission: April 10, 2018
Submission of the description of methods: April 15, 2018
Conference: August 20-24, 2018
More Information and Registration
Contact
Organizers
Chloé Artaud (L3i, University of La Rochelle, France)
Nicolas Sidère (L3i, University of La Rochelle, France)
Antoine Doucet (L3i, University of La Rochelle, France)
Jean-Marc Ogier (L3i, University of La Rochelle, France)
Vincent Poulain d’Andecy (Yooz, Aimargues, France)
Reply all
Reply to author
Forward
0 new messages