Re: [Gimp-developer] A GSoC Purpose: Improve the Performance of SIOX
Hi, On Sat, 2009-03-14 at 11:52 +0800, Jenny wrote: My name is Ding Jie, and Jenny is my English name. I'm a first-grade post-graduate student, major in Image Processing, and very interested in algorithms of this field. Thanks for neo's suggestion, I found SIOX is very interesting. I have read some paper about SIOX written by Gerald, and feel SIOX have potential to improve. Firstly, SIOX performs not good when dealing with gray images. Secondly, it isn't good at extracting things from a background whose color is similar to foreground. I'm wondering is it possible to use some new approaches side-by-side, and make the final output as the combination of all those approaches' output. SIOX can by definition only extract things from a background if the foreground color is sufficiently different. After all that is how the algorithm decides what to separate. So what you suggest here sounds more like adding a different separation algorithm and not like improving SIOX. But of course I might just have misunderstood you. The SIOX tool has several areas where it is lacking. The most notable is that it does a binary separation. Either a pixel is foreground or it is background. This causes jaggy edges and makes it necessary to do manual post-processing on the selection. The SIOX team around Gerald has suggested several improvements that would help to improve this. Their papers describe a refinement brush that should be used on the selection boundary to improve the results. It would be nice if this could be added to the GIMP SIOX tool. Sven ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] A GSoC Purpose: Improve the Performance of SIOX
Hi, On Sat, Mar 14, 2009 at 6:50 PM, Sven Neumann s...@gimp.org wrote: SIOX can by definition only extract things from a background if the foreground color is sufficiently different. After all that is how the algorithm decides what to separate. So what you suggest here sounds more like adding a different separation algorithm and not like improving SIOX. But of course I might just have misunderstood you. The SIOX tool has several areas where it is lacking. The most notable is that it does a binary separation. Either a pixel is foreground or it is background. This causes jaggy edges and makes it necessary to do manual post-processing on the selection. The SIOX team around Gerald has suggested several improvements that would help to improve this. Their papers describe a refinement brush that should be used on the selection boundary to improve the results. It would be nice if this could be added to the GIMP SIOX tool. Thank you for your feedback. I have only read some papers about how SIOX extract thing, and didn't sure the real lack point of SIOX. Your suggestion is very helpful. :) I didn't read the code of SIOX and GIMP yet, and be not sure if the features in this page http://www.siox.org/preview.html has been implemented in GIMP's current development version. If not, is it a good a GSoC project of implementing these features? And I'm still not sure if Gerald want to be a mentor in this GSoC Season. BTW, I have idled on the IRC channel #gimp for days, where I'm glad if anyone could give me suggestions, or just talk :) Cheers, Jenny ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] A GSoC Purpose: Improve the Performance of SIOX
Hi, On Sat, 2009-03-14 at 19:41 +0800, Jenny wrote: I didn't read the code of SIOX and GIMP yet, and be not sure if the features in this page http://www.siox.org/preview.html has been implemented in GIMP's current development version. The Detail Refinement Brush that is mentioned on this page has not been implemented yet. And I'm still not sure if Gerald want to be a mentor in this GSoC Season. I have forwarded your mail to him and asked him if he'd be interested to help us having a GSoC project on SIOX. Please give him some time to respond. He is probably quite busy in his new job at the International Computer Science Institute (ICSI) in Berkeley. Sven ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] A GSoC Purpose: Improve the Performance of SIOX
Hi, Jenny, nice to meet you! I agree with Sven that implementing the Detail Refinement Brush as described on the http://siox.org website would be something that could be done immediately, i.e. without performing extensive computer vision research and would contribute significantly to the improvement of the SIOX tool. Once it is done, there is still room for research, e.g. how to automate the detail refinement brush further and also HCI issues. I am glad to respond to any questions by email and could be a mentor as far as conceptual questions (as opposed to GIMP code structure) are concerned. Gerald On Sat, Mar 14, 2009 at 5:05 AM, Sven Neumann s...@gimp.org wrote: Hi, On Sat, 2009-03-14 at 19:41 +0800, Jenny wrote: I didn't read the code of SIOX and GIMP yet, and be not sure if the features in this page http://www.siox.org/preview.html has been implemented in GIMP's current development version. The Detail Refinement Brush that is mentioned on this page has not been implemented yet. And I'm still not sure if Gerald want to be a mentor in this GSoC Season. I have forwarded your mail to him and asked him if he'd be interested to help us having a GSoC project on SIOX. Please give him some time to respond. He is probably quite busy in his new job at the International Computer Science Institute (ICSI) in Berkeley. Sven ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer -- Dr. Gerald Friedland International Computer Science Institute 1947 Center Street, Suite 600 CA-94704 Berkeley, USA Office: +1/510/666-2987 Mobile: +1/510/529-6514 http://www.gerald-friedland.org ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
