ICFHR 2014
Handwritten KeyWord Spotting Competition
Handwritten keyword spotting is the task of detecting query words in handwritten document image collections without involving a traditional OCR step. Recently, handwritten word spotting has attracted the attention of the research community in the field of document image analysis and recognition since it appears to be a feasible solution for indexing and retrieval of handwritten documents in the case that OCR-based methods fail to deliver satisfactory results. In the proposed competition, we plan to set up an evaluation framework for benchmarking handwritten keyword spotting approaches which address the query by example problem. The task considered for evaluation will be as follows. A query word image along with a collection of handwritten document images will be provided as input to the system under evaluation. The expected output will be a ranked list of bounding boxes, which correspond to spotted word images that match the query word image in terms of a similarity value.
In Handwritten KeyWord Spotting Competition 2014 (H-KWS 2014), the main objective is to record current advances in keyword spotting using established evaluation performance measures frequently encountered in the multimedia retrieval literature. The competition will comprise two distinct tracks, namely, a segmentation-based and a segmentation-free track. The benchmarking dataset that will be used in the contest will contain both historical and modern documents from multiple writers.
All researchers in the field of Document Image Analysis that put emphasis on keyword spotting are invited to register for participating in H-KWS 2014. Interested parties should pay attention at two critical milestones. The first (April 1, 2014) is the release of the dataset along with the query set to the registered participants and the second (April 7, 2014) is the required outcome with the produced ranked lists delivered by the participants. Detailed information about the structured input / output will be given after registration. For H-KWS 2014, binary executables are not required.
The description of the participating methods, the evaluation methodology and the final ranking will be presented during a dedicated ICFHR 2014 session. A report on the competition will be published in the ICFHR 2014 conference proceedings. At the end of the competition, the testing dataset along with the required software to run the evaluation benchmark will become publicly available.
NOTE: You may participate in this contest even if you do not plan to attend the ICFHR 2014 conference.
Important Dates.
04/09/2014 - Final contest results to be announced at the ICFHR 2014.
