What's new?
Deep OCR Automated Labeling: The new Automated Labeling feature for Deep OCR allows you to configure detection and recognition parameters to automatically generate label proposals. You can quickly accept or adjust these proposals, significantly speeding up the labeling process for large datasets and reducing the need for manual intervention.
Integrated Detection and Recognition Training for Deep OCR: With version 24.12, training both detection and recognition networks is now possible within a single tool. This integration simplifies the training workflow, enabling you to develop and fine-tune your models more efficiently without switching between different applications.
Smart Segmentation Labeling 2.0: Smart Segmentation Labeling 2.0 offers improved accuracy and faster label suggestions for segmentation tasks. Additionally, the preprocessing time for each image has been reduced, allowing for smoother and quicker labeling sessions even on less powerful hardware.
How to get it?
You can download and install this new version via the MVTec Software Manager (SOM). Be sure to first hit “refresh from server” to get the latest list of available versions. For users who cannot use the online installer, we also provide an offline installer on the MVTec website.