CPC G06V 10/25 (2022.01) [G06F 18/214 (2023.01); G06F 18/217 (2023.01); G06V 10/34 (2022.01); G06V 10/774 (2022.01); G06V 20/653 (2022.01); G06V 2201/10 (2022.01)] | 20 Claims |
1. A method comprising:
capturing image data of a product using a plurality of cameras, the image data including a plurality of image frames and being captured while the product is moving in one or more dimensions;
associating a product identifier for the product with each of the plurality of image frames;
identifying a plurality of segmentation masks comprising a segmentation mask for the product in each of the plurality of image frames each segmentation mask corresponding to an outline of the product as derived from a respective bounding box identified in each of the plurality of image frames; and
optimizing the plurality of segmentation masks by inputting the plurality of image frames into a machine learning model, the machine learning model trained on the plurality of segmentation masks, the machine learning model outputting a set of optimized segmentation masks for the product.
|