evaluation of problems with the model at KITTI2015

[JF-Lee]

Hi, I try to use torchvision.utils.save_image() to save the output tensor of the test, but got these blank images, and I try to convert tensor to PIL format, the image is still blurry. So, how can I get the real disp image? I hope to get your answer. Thanks!

[RayneSun]

https://github.com/mileyan/AnyNet/issues/12#issuecomment-638191181

[JF-Lee]

The code in the picture can help you！

------------------ 原始邮件 ------------------ 发件人: "RayneSun"<[email protected]>; 发送时间: 2020年6月15日(星期一) 上午8:40 收件人: "mileyan/AnyNet"<[email protected]>; 抄送: "走走停停ゝ"<[email protected]>;"Author"<[email protected]>; 主题: Re: [mileyan/AnyNet] evaluation of problems with the model at KITTI2015 (#19)

Oh, I met the same matter! I use pretrained network to output a picture from kitti 2015, but got a very noisy picture! image

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.