π·Retrain Mode
If you have imported a neural render scanned through the Incarnate app, you can retrain the model using the Retrain mode, which offers two essential functionalities:
Image Visualiser
The Image Visualiser allows you to view each image captured during the product scanning process in the Incarnate app. These images are displayed from the exact angles at which they were originally captured. You have the option to remove any unwanted images by clicking on them and utilizing the 'Remove Image Button.'
Bounding Box Realignment
Bounding Box Realignment enables you to correct the placement of the bounding box if it was
inaccurately positioned during the product scan through the Incarnate app. When you access this option, you can observe the precise positioning of the bounding box relative to the scanned
product. You also have the ability to adjust the size of the bounding box along the X, Y, and Z axes using the options available in the right-side panel.
Once you have made the necessary changes in the Image Visualiser and Bounding Box Realignment sections, simply click on the 'Retrain' button and patiently wait for the AI to complete the remaining tasks. Soon, you can expect a significantly improved output that aligns with your expectations.
Important
This option is only available if you are using a Neural render (scanned model through Incarnate by Avataar) and wonβt be visible if you are using an uploaded 3D model.
Last updated