horizontal-text-detection-0001

Use Case and High-Level Description

Text detector based on FCOS architecture with MobileNetV2-like as a backbone for indoor/outdoor scenes with more or less horizontal text.

The key benefit of this model compared to the base model is its smaller size and faster performance.

Metric	Value
F-measure (harmonic mean of precision and recall on ICDAR2013)	88.45%
GFlops	7.78
MParams	2.26
Source framework	PyTorch*

Image, name: input, shape: 1, 3, 704, 704 in the format 1, C, H, W, where:

Expected color order - BGR.

The boxes is a blob with the shape 100, 5 in the format N, 5, where N is the number of detected bounding boxes. For each detection, the description has the format: [x_min, y_min, x_max, y_max, conf], where:
- (x_min, y_min) - coordinates of the top left bounding box corner
- (x_max, y_max) - coordinates of the bottom right bounding box corner
- conf - confidence for the predicted class
The labels is a blob with the shape 100 in the format N, where N is the number of detected bounding boxes. In case of text detection, it is equal to 0 for each detected box.

The OpenVINO Training Extensions provide a training pipeline, allowing to fine-tune the model on custom dataset.

[*] Other names and brands may be claimed as the property of others.