ICCV Daily 2021 - Wednesday

which encourage closer appearances to the input image by comparison to previous works. ” Xiangyu hopes to extend the Texformer architecture to more advanced human models. He also believes it could be generalized to many other tasks. The key to the success of this work is its ability to interchange information between different spaces, which could be generalized to other tasks that require information interchange between different spaces, such as multi-modality learning and graph processing. Originally from China , Xiangyu is currently enjoying his time at Nanyang Technological University in Singapore . “ It is a very nice university located in beautiful Singapore, ” he muses. “ This place is very diverse. There are people from all different countries and races here. It really is a great atmosphere! ” To learn more about Xiangyu’s work [Paper ID 1281], you can view his pre- recorded oral presentation and PDF poster now and join his live Q&A sessions today [11A] at 18:00-19:00 EDT and Friday [11B] at 11:00-12:00 EDT. 7 DAILY ICCV Wednesday Xiangyu Xu Wait, before you continue, did you check my picks for today on page 2?