Transformer-Based Attention Networks for Continuous Pixel-Wise Prediction
        Guanglei Yang1,2 Hao Tang3 Mingli Ding1 Nicu Sebe2 Elisa Ricci2,4
        1
         Harbin Institute of Technology, China 2 DISI, University of Trento, Italy
      3
       Computer Vision Lab, ETH Zurich, Switzerland 4 Fondazione Bruno Kessler, Italy
{yangguanglei,dingml}@hit.edu.cn,hao.tang@vision.ee.ethz.ch,{niculae.sebe,e.ricci}@unitn.it

             Abstract                  to improve their ability in mode ...