Box2Poly: Memory-Efficient Polygon Prediction of Arbitrarily Shaped and Rotated Text
DOI:
https://doi.org/10.1609/aaai.v38i2.27884Keywords:
CV: Object Detection & Categorization, CV: Scene Analysis & UnderstandingAbstract
Recently, Transformer-based text detection techniques have sought to predict polygons by encoding the coordinates of individual boundary vertices using distinct query features. However, this approach incurs a significant memory overhead and struggles to effectively capture the intricate relationships between vertices belonging to the same instance. Consequently, irregular text layouts often lead to the prediction of outlined vertices, diminishing the quality of results. To address these challenges, we present an innovative approach rooted in Sparse R-CNN: a cascade decoding pipeline for polygon prediction. Our method ensures precision by iteratively refining polygon predictions, considering both the scale and location of preceding results. Leveraging this stabilized regression pipeline, even employing just a single feature vector to guide polygon instance regression yields promising detection results. Simultaneously, the leverage of instance-level feature proposal substantially enhances memory efficiency ( > 50% less vs. the SOTA method DPText-DETR) and reduces inference speed (> 40% less vs. DPText-DETR) with comparable performance on benchmarks. The code is available at https://github.com/Albertchen98/Box2Poly.git.Downloads
Published
2024-03-24
How to Cite
Chen, X., Wang, D., Schindler, K., Sun, M., Wang, Y., Savioli, N., & Meng, L. (2024). Box2Poly: Memory-Efficient Polygon Prediction of Arbitrarily Shaped and Rotated Text. Proceedings of the AAAI Conference on Artificial Intelligence, 38(2), 1219-1227. https://doi.org/10.1609/aaai.v38i2.27884
Issue
Section
AAAI Technical Track on Computer Vision I