Real-time Lexicon-free Scene Text Retrieval
| Autor: | Marçal Rusiñol, Andres Mafla, Lluis Gomez, Ernest Valveny, Sounak Dey, Rubèn Tito, Dimosthenis Karatzas |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Generalization
business.industry Computer science Nearest neighbor search 02 engineering and technology Spotting computer.software_genre Lexicon 01 natural sciences Task (computing) Artificial Intelligence 0103 physical sciences Signal Processing 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Computer Vision and Pattern Recognition Artificial intelligence 010306 general physics Representation (mathematics) business computer Software Natural language processing Text retrieval |
| Zdroj: | Pattern Recognition. 110:107656 |
| ISSN: | 0031-3203 |
| DOI: | 10.1016/j.patcog.2020.107656 |
| Popis: | In this work, we address the task of scene text retrieval: given a text query, the system returns all images containing the queried text. The proposed model uses a single shot CNN architecture that predicts bounding boxes and builds a compact representation of spotted words. In this way, this problem can be modeled as a nearest neighbor search of the textual representation of a query over the outputs of the CNN collected from the totality of an image database. Our experiments demonstrate that the proposed model outperforms previous state-of-the-art, while offering a significant increase in processing speed and unmatched expressiveness with samples never seen at training time. Several experiments to assess the generalization capability of the model are conducted in a multilingual dataset, as well as an application of real-time text spotting in videos. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect