SWAG-V: Explanations for Video using Superpixels Weighted by Average Gradients.

Thomas Hartley Kirill A. Sidorov Christopher Willis A. David Marshall

Published in: WACV (2022)

Keyphrases

superpixels
multiscale
natural images
input image
object segmentation
segmentation algorithm
video frames
long range
perceptual grouping
segmentation method
bounding box
video data
image patches
video sequences
conditional random fields
multiple images
graph structure
multi class
contour fragments