Guided Visual Attention Model Based on Interactions Between Top-down and Bottom-up Information for Robot Pose Prediction
Guided Visual Attention Model Based on Interactions Between Top-down and Bottom-up Information for Robot Pose Prediction
Hyogo Hiruma,Hiroki Mori,T. Ogata
2022 · DBLP: journals/corr/abs-2202-10036
arXiv.org · 2 Citations
TLDR
A novel Key-Query-Value formulated visual attention model which can be guided to a certain attention target is proposed, showing better performance compared to existing end-to-end robot vision models and its high scalability and extendiblity on robot controlling tasks.