Skip to content

istiakshihab/automated-retail-checkout-aicity22

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

VISTA: Vision Transformer enhanced by U-Net and Image Colorfulness Frame Filtration for Automatic Retail Checkout

Code for our CVPR 2022 Workshop paper VISTA: Vision Transformer enhanced by U-Net and Image Colorfulness Frame Filtration for Automatic Retail Checkout. The method described achieves 3rd place in the AI City Challenge 2022 Track 4: Multi-Class Product Counting & Recognition for Automated Retail Checkout. See here for details.

[arXiv]


Figure 1. Illustration of the overall segmentation and classification pipeline

1. Specification of dependencies

This code requires Python 3.8.12 and PyTorch 1.8.2. Run pip install -r requirements.txt to install all the dependencies.

2a. Segmentation training code

See training/segmentation for details.

2b. Classification training code

See training/classification for details.

3. Inference code

After steps 2a and 2b, make sure both segmentation and classification models are present in the test/models directory. Then see README.md for details.

4. Citation

will be added here.

Acknowledgements

We thank AICITY 22 organizers for making data available for use. We also thank Giga Tech Ltd. for providing funding for this work.