S3DIS

[burui11087]

Thanks for your great work!!!!

I have a question about S3DIS. As we known, SEGColud just train on area 1-4,and 6, test on area5. However, others use six cross folder train/test. So the results are in different experiment environments, I think we shouldn't compare them together.

BTW, I am looking forward to your S3DIS code release.

Thanks!!

[jmydurant]

OK, I am looking for internship right now. I will release it after that.

https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&name=%E4%BF%AE%E5%AE%87%E4%BA%AE&uid=example%40163.com&ftlId=3&items=%5B%22%E9%82%AE%E7%AE%B1%EF%BC%9Ajmydurant%40hotmail.com%22%5D [https://mail-online.nosdn.127.net/qiyelogo/defaultAvatar.png] 修宇亮 邮箱：[email protected]

签名由 网易邮箱大师https://mail.163.com/dashi/dlpro.html?from=mail88 定制

On 07/19/2018 14:03, Rui Bumailto:[email protected] wrote:

Thanks for your great work!!!!

I have a question about S3DIS. As we known, SEGColud just train on area 1-4,and 6, test on area5. However, others use six cross folder train/test. So the results are in different experiment environments, I think we shouldn't compare them together.

BTW, I am looking forward to your S3DIS code release.

Thanks!!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://nam03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FMVIG-SJTU%2FpointSIFT%2Fissues%2F8&data=02%7C01%7C%7C3edbbe7f625c4b708e8408d5ed3d58b3%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636675770084065213&sdata=apPaEuYp55NwGZRdYe8IFikXO7sBLXn0DAznwWBwxUw%3D&reserved=0, or mute the threadhttps://nam03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAM1xfJOU1Yf-ZX-4VGvpOSZYD44-JBAFks5uICEvgaJpZM4VVzC1&data=02%7C01%7C%7C3edbbe7f625c4b708e8408d5ed3d58b3%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636675770084065213&sdata=%2FNFHwsh1Qc3uLhYxmy6tMd0h%2FuGViw0zxUzGkKM6l%2Fk%3D&reserved=0.

[LZDSJTU]

I have a question on Scannet dataset.

In your code, you use 1201 scenes in training and 312 in validation. So the final accuracy in your paper is the validation result instead of your test result? I find that the test.py is empty, and I want to know how you calculate the mean IOU and Overall accuracy in your paper.

Thank you very much.

[yangxh31]

i have the same problems with scannet, and i also want to know what's the format of the dataset of scannet, the reason why i want to train my own dataset and don't know how to change the format.

thank you very much!!!