FeedFormer: Revisiting Transformer Decoder for Efficient Semantic Segmentation.

Jae-hun Shim Hyunwoo Yu Kyeongbo Kong Suk-Ju Kang

Published in: AAAI (2023)

Keyphrases

semantic segmentation
superpixels
weakly supervised
street scenes
information extraction
image representation
conditional random fields
biologically inspired
scene classification