Attention-based Extraction of Structured Information from Street View Imagery

11 Apr 2017 • Zbigniew Wojna • Alex Gorban • Dar-Shyang Lee • Kevin Murphy • Qian Yu • Yeqing Li • Julian Ibarz

We present a neural network model - based on CNNs, RNNs and a novel attention mechanism - which achieves 84.2% accuracy on the challenging French Street Name Signs (FSNS) dataset, significantly outperforming the previous state of the art (Smith'16), which achieved 72.46%. Furthermore, our new method is much simpler and more general than the previous approach... (read more)

PDF Abstract

Tasks
Add Remove

Datasets

COCO-Text
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Optical Character Recognition FSNS - Test AttentionOCR_Inception-resnet-v2_Location Sequence error 15.8 # 1
Compare

Methods used in the Paper
Add Remove

METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet
Contact us on: hello@paperswithcode.com. Papers With Code is a free resource with all data licensed under CC-BY-SA.
Terms Privacy Cookies policy