question about the superpixel segmentation

Question

question about the superpixel segmentation

colorfuloranges opened this issue 2 years ago · comments

Thanks for your sharing! It's very helpful to my research. I have a question about the code. The JPSA is a two-stream multilayered regression model involving the two input sources: 1) pixelwise and 2) superpixelwise spectral signatures and the same output(ground truth). But l do not see the input of the supempixel segmentation part and the superpixel segmentation of images in the code. l dont know if l understand the meaning of the paper correctly?Looking forward to your reply, thank you！

Danfeng Hong · Answer 1 · Mon Aug 01 2022 19:15:53 GMT+0800 (China Standard Time)

The code is its previous version (pixel-wise version, i.e., ECCV2018) colorfuloranges ***@***.***> 于2022年8月1日周一 16:14写道：

…

Thanks for your sharing! It's very helpful to my research. I have a question about the code. The JPSA is a two-stream multilayered regression model involving the two input sources: 1) pixelwise and 2) superpixelwise spectral signatures and the same output(ground truth). But l do not see the input of the supempixel segmentation part and the superpixel segmentation of images in the code. l dont know if l understand the meaning of the paper correctly?Looking forward to your reply, thank you！ — Reply to this email directly, view it on GitHub <#2>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFL2GZUVH7VNVQETITSOSZTVW6BP3ANCNFSM55GUEJEQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

colorfuloranges · Answer 2 · Mon Aug 01 2022 19:40:39 GMT+0800 (China Standard Time)

Thank you very much for your reply. If I want to reproduce this algorithm, can the average spectral features extracted after superpixel be directly concatenated after pixel-level spectral features for input?

Danfeng Hong · Answer 3 · Mon Aug 01 2022 19:50:33 GMT+0800 (China Standard Time)

yes, your idea is promising and working well, I think. You can try it, if it works well, please also let me know at your convenience! Thanks! Danfeng colorfuloranges ***@***.***> 于2022年8月1日周一 19:40写道：

…

Thank you very much for your reply. If I want to reproduce this algorithm, can the average spectral features extracted after superpixel be directly concatenated after pixel-level spectral features for input? — Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFL2GZV4IV6ICU564VGOMYLVW6ZUDANCNFSM55GUEJEQ> . You are receiving this because you commented.Message ID: ***@***.***>

colorfuloranges · Answer 4 · Mon Aug 01 2022 20:56:48 GMT+0800 (China Standard Time)

Thank you for your encouragement. I'll try it. Thank you very much!