PlaceFormer: Transformer-based Visual Place Recognition using Multi-Scale Patch Selection and Fusion.

Shyam Sundar Kannan Byung-Cheol Min

Published in: CoRR (2024)

Keyphrases

vision system
multiscale
visual information
image fusion
visual features
fuzzy logic
visual perception
image patches
power system
data fusion
image processing
multi sensor
scale space
coarse to fine
low level
visual data
visual cues
wavelet transform
human vision
fine scale