Mindee has announced the introduction of docTR, a seamless, high-performing, and accessible open-source library for OCR-related tasks powered by deep learning.

Mindee’s docTR provides optical character recognition with accessibility for the entire developer community. Combining textual parsing through text and object detection and recognition, this open-source repository offers a wider range and complex use cases.

Going beyond the textual elements, it provides a holistic view of information encoded in visual forms, including QR codes, barcodes, information in ID pictures, and even logos.

Powered by the machine learning tool of your choice, TensorFlow 2 or PyTorch, DocTR features training capabilities for text detection in documents and images as well as recognition with pretrained parameters.

It incorporates a five-line code to load documents, extract text with a predictor, and optimise for very high end-to-end performances, including inference speed on both CPU and GPU.

With this offering, Mindee provides a wide audience, from entry-level developers to domain experts who want to train their model (researchers), the tools to support efforts in their transformation from intensive manual data entry (e.g., from physical documents, PDFs or images) to a full digital process.

docTR was developed to provide organisations with tangible results ranging from time savings through the development process; easy integration with existing systems and architectures; minimized deployment costs; to increased productivity across departments with faster retrieval of information from documents.

“Releasing docTR as an open-source library opened a world of possibilities for innovation,” said Frédéric Harper, Director of Developer Relations at Mindee.

“At Mindee, we take pride in adding value to the developer community. We made this code available with that in mind, to ensure developers can read it, understand it and be sure it’s safe. We are providing everyone with the possibility of making this OCR tool their own by allowing them to modify the code to fit their applications and infrastructure needs.”

docTR is fully available now with multiple ways to access:

• Install it as a Python library using ‘pip install python-doctr’, here: python-doctr 0.4.1
• Download the tool directly from the GitHub repository here: https://github.com/mindee/doctr/releases
• Use the source to compile it yourself here: https://github.com/mindee/doctr

To learn more about Mindee, please visit Mindee.com