TY  - JOUR
AU  - Andres Mafla
AU  - Ruben Tito
AU  - Sounak Dey
AU  - Lluis Gomez
AU  - Marçal Rusiñol
AU  - Ernest Valveny
AU  - Dimosthenis Karatzas
PY  - 2021//
TI  - Real-time Lexicon-free Scene Text Retrieval
T2  - PR
JO  - Pattern Recognition
SP  - 107656
VL  - 110
N2  - 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.
UR  - https://www.sciencedirect.com/science/article/pii/S0031320320304593
N1  - DAG; 600.121; 600.129; 601.338
ID  - Andres Mafla2021
ER  -