US 12,347,222 B2
Visualization of the impact of training data using bounding box vectors to train a machine learning model
Zhong Fang Yuan, Xi'an (CN); Tong Liu, Xi'an (CN); Pitipong Jun Sen Lin, Cambridge, MA (US); Elaine Marie Branagh, Austin, TX (US); and Chen Yu Chang, Xi'an (CN)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Jan. 27, 2022, as Appl. No. 17/586,343.
Prior Publication US 2023/0237827 A1, Jul. 27, 2023
Int. Cl. G06V 30/414 (2022.01); G06V 30/19 (2022.01); G06V 30/412 (2022.01); G06V 30/418 (2022.01); G06V 10/82 (2022.01)
CPC G06V 30/412 (2022.01) [G06V 30/19013 (2022.01); G06V 30/414 (2022.01); G06V 30/418 (2022.01)] 18 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
a processor that, when executing instructions stored in a memory, is configured to:
generate a plurality of bounding boxes at a plurality of content areas in an image, wherein the plurality of bounding boxes correspond to a plurality of pieces of text within the image;
convert the plurality of bounding boxes into a plurality of bounding box vectors based on attributes of the plurality of bounding boxes;
select a bounding box of the plurality of bounding boxes as an anchor bounding box;
identify bounding boxes that are adjacent to the anchor bounding box;
concatenate bounding box vectors corresponding to the bounding boxes, that are adjacent to the anchor bounding box, into a concatenated bounding box vector;
store a mapping between the anchor bounding box and the concatenated bounding box vector;
train a machine learning model to create a trained machine learning model to transform a bounding box into a location in vector space based on mappings, including the concatenated bounding box vector; and
store the machine learning model in the memory.