VISTA: Vision Transformer enhanced by U-Net and Image Colorfulness Frame Filtration for Automatic Retail Checkout.
Md. Istiak Hossain ShihabNazia TasnimHasib ZunairLabiba Kanij RuptyNabeel MohammedPublished in: CVPR Workshops (2022)
Keyphrases
- input image
- image features
- image classification
- image data
- image frames
- template matching
- image content
- low level image processing
- vision system
- image analysis
- multiscale
- image segmentation
- edge detection
- feature points
- single image
- region of interest
- image collections
- coordinate frame
- image representation
- image regions
- image retrieval
- hough transform
- single frame
- image processing
- image pixels
- low level
- test images
- fuzzy logic
- spatial information
- high resolution
- key frames
- reference frame
- visual perception
- real time
- computer vision
- scan line
- neural network