Picterra incorporates testing areas into workflow
Information from Picterra
Picterra has released a new feature, Testing Areas, which is now an integral part of the workflow for training a detector. Its purpose is to visually assess the results while building the detector. Users now need to follow three steps in training mode:
- Defining training areas
- Drawing object outlines within the defined training areas
- Defining testing areas
The testing areas was made part of building the detector since it makes it possible to quickly grasp where the detector needs improving across possibly very visually different images. Previously, it was only possible to assess the training results on one image at a time. Testing Areas make it possible to control exactly where the model preview outputs during training. The feature also speeds up the training iterations, as previously, especially on big images, generating detection across the entire image used to take more time.
Testing areas lets users focus exactly on the parts of the image that are representative of the whole set and allows them to visually assess the detector performance.