ICCV：Weakly Supervised Relative Spatial Reasoning for Visual Question Answering

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

1908
1909
monocular (single-image) depth estimation remains a chal-       geometry-based training objectives – object centroid esti-
lenging problem, with learning-based methods pushing the       mation (OCE) and relative position estimation (RPE). In
envelope [43, 11, 31]. In this work, we utilize AdaBins [6]      this section, we describe these SR tasks.
which uses a transformer-based architecture that adaptively
divides depth ranges into variable-sized bins and estimates      3.1. Pre-Proce ...

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Reasoning answering Relative question Spatial