Name
Abstract | ||
|---|---|---|
Word spotting systems are intended to retrieve occurrences of a given keyword in document images without actually recognizing the full document content. As there is a trend towards segmentation-free word spotting methods, we propose a methodology to evaluate these methods by employing measures that take the quality of the retrieved word locations into account without making hard decisions. We derive a desired evaluation behavior with the help of synthetic examples and show discrepancies of existing evaluation methods. New measures following this behavior are introduced and their differences exemplarily described. The proposed evaluation method is applied to a state-of-the-art word spotting approach. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/ICDAR.2013.263 | ICDAR-1 |
Keywords | Field | DocType |
document image processing,image retrieval,document images,keyword occurrence retrieval,segmentation-free word spotting approach evaluation,word location retrieval quality,Evaluation,Location Quality,Measure,Precision,Recall,Segmentation-Free,Word Spotting | Pattern recognition,Segmentation,Computer science,Document image processing,Image retrieval,Speech recognition,Artificial intelligence,Recall,Spotting,Visual Word | Conference |
ISSN | Citations | PageRank |
1520-5363 | 1 | 0.42 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Werner Pantke | 1 | 1 | 0.42 |
| Volker Märgner | 2 | 295 | 29.02 |
| Tim Fingscheidt | 3 | 164 | 48.89 |