foxkw

> HI. thanks for your great job. > > I found that warped_detector_ Loss is difficult to converge when training superpoint . because the image content changes with each epoch,...

> Hi, ideally, both the source and target image should be warped during training, to improve the diversity of the training images. I think that currently, when training on pairs...

> Hi, the prediction will never be perfect anyway, and the current one seems pretty good in my opinion. Since the green point can cover the red ones, it is...

> > you can write a script to transform the format. 发自我的华为手机-------- 原始邮件 --------发件人： Ahmed Nasr Ahmed 日期： 2020年5月16日周六 早上7:01收件人： TuSimple/tusimple-benchmark 抄送： shijia-web , Comment 主 题： Re: [TuSimple/tusimple-benchmark] How...

> As same as segment (instance segmentation), using polygon & add additional line to link polygons if the stuff category has multiple parts. Hi! Regarding the panoptic segmentation task, where...