Case Study: Retail
Contactless Checkout is the new retail mantra for increased satisfaction
Increased footfalls in retail stores ends up in long queues at check-out. AI and Cognitive expertise can help solve this problem especially for large retail chains.
A large retail chain based in Hong Kong uses our cognitive expertise across 350 retail outlets to reduce check-out time for their customers.
Abzooba implemented end to end flow of contactless checkout solution by the following steps:
- Labeling: Have SKU/instance specific bounding box annotations on the captured (and/or synthetic) scene images
- Training: Learn a deep neural net to detect the correct SKU instances from a scene
- Object detection using tensorflow for 10 classes
- Faster-RCNN with ResNet101 based model
- Integrated with desktop demo
- Accuracy: 93% mean Precision, 93% mIoU
- Inference model: Publish the best performing model on cloud/premise/device
- UPLOAD image captured via POST request
- Get JSON response – list of SKU/barcodes
- Communicate with billing software
- Share barcodes/SKU info to generate bill on the billing software
Abzooba developed Computer Vision object detection model with 93% mean precision. The client observe d a 75% reduction in checkout time, from ~6 seconds to ~1.5 seconds.