RGBD-SODsurvey
RGBD-SODsurvey copied to clipboard
taozh2017
Custom dataset
Hello, Thank you for sharing your work. Can you provide the different custom RGB-D dataset (indoor vs outdoor, object scale, background clutter, ...) ?
I have updated it on the readme page:
Attribute-based Evaluation: Downloading attribute datasets Baidu Drive (Code: tktz) or Google Drive.
Besides, for scale evaluation, just using: obj_size = length(find(gt~=0)) / (size(gt,1)*size(gt,2)), to compute metrics for different sacles.
Thank you. Just one more question. When you did the training, you trained each model once on a dataset of randomly selected samples (700 from NLPR, 1500 from NJU2K) ? For the testing, you just used the custom datasets without retraining the models right ?
Yes, without retraining.
y0un0 @.***> 于2021年11月10日周三 下午11:34写道：
Thank you. Just one more question. When you did the training, you trained each model once on a dataset of randomly selected samples (700 from NLPR, 1500 from NJU2K) ? For the testing, you just used the custom datasets without retraining the models right ?
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/taozh2017/RGBD-SODsurvey/issues/5#issuecomment-965417364, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHT5QRL6EJO4IJZX4FHIIZTULKGJTANCNFSM5HKG3J4A . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
-- Best,
Tao.