Re: [Gimp-developer] Rotational Motion Blur
On 08/18/10 19:09, yahvuu wrote: On 18.08.2010 12:50, g...@catking.net wrote: On 08/18/10 11:07, Tor Lillqvist wrote: A motion blur is a retinal effect that has a time dependence. Is motion blur actually something people perceive with their eyes and brain, or something that only exists in physical artefacts? (Either intentionally created by an artist to give the impression of motion, or as an direct result of the method the still or motion picture was created.) And we have been so used to it that we know what it means, even if it doesn't correspond to what we actually see? (But yeah, gg's arguments make sense.) --tml Good point, the equal weighting probably is close to what a silver nitrate film camera would record I think so, too. Consider the star trails in a long-time exposure of a night sky: there is no decay visible. However, and whatever the motivation, it's an interesting idea, so here's a quick comparison for a linear motion blur: http://yahvuu.files.wordpress.com/2009/08/blurtest.png Regretably, the mathmap convolve function introduces some artifacts, but i think it can be seen that 'decaying' (or 'soft'?) motion blur is an option of artistical relevance. Tongue in cheek, i shurely wouldn't oppose if someone wanted to get code providing this functionality included into GIMP .- regards, yahvuu The blurtest.png is interesting. The example with the decaying kernel is the only one that suggests movement to me , it definitely has direction. All the others look, well, blurred. Despite the artifacts of the implementation, this suggests some sort of progressive blur may be interesting and may suggest movement better than current photographic blur. /gg ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
On 08/18/10 01:20, Bill Skaggs wrote: I just looked over the code -- if I understand it correctly, it creates an arc of the specified angle passing through each point of the image, generates a set of points equally spaced along the arc, uses interpolation to get the color for each of these points, and then averages all the colors together, with equal weighting. The number of points is proportional to the length of the arc, with adjustments if it is too large or too small. If you need more detail, please ask. -- Bill I'm sure you are right in reading the code, so this feature seems more like a rotation blur than motion blur. A motion blur is a retinal effect that has a time dependence. This does not seem to be what is done. The equal weighting is wrong. The retinal retention of the image fades with time (my guess would be exponentially if we're going to be rigorous). This means the weighting of each point should be diminished as we work back along the arc. This reduction should be angular and at least a linear reduction of weight w.r.t. the angular coord would probably be a decent approximation. Otherwise a lookup table of exponential decay would mean very little overhead in doing a fairly rigorous model of the real effect. This effect will vary dependant on speed of the movement (as everyone knows intuitively) so this parameter needs to be adjustable in the tool UI. I assume the same points could be made about the motion blur , though I have not looked at the code. I'd don't have the time to get into coding an improvement but thought it worth noting the apparent shortcomings of the motion blur and how it could be improved. regards. ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
A motion blur is a retinal effect that has a time dependence. Is motion blur actually something people perceive with their eyes and brain, or something that only exists in physical artefacts? (Either intentionally created by an artist to give the impression of motion, or as an direct result of the method the still or motion picture was created.) And we have been so used to it that we know what it means, even if it doesn't correspond to what we actually see? (But yeah, gg's arguments make sense.) --tml ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
On 08/18/10 11:07, Tor Lillqvist wrote: A motion blur is a retinal effect that has a time dependence. Is motion blur actually something people perceive with their eyes and brain, or something that only exists in physical artefacts? (Either intentionally created by an artist to give the impression of motion, or as an direct result of the method the still or motion picture was created.) And we have been so used to it that we know what it means, even if it doesn't correspond to what we actually see? (But yeah, gg's arguments make sense.) --tml Good point, the equal weighting probably is close to what a silver nitrate film camera would record, which is probably what this was intended to model. /gg ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
On Wed, Aug 18, 2010 at 12:50:34PM +0200, g...@catking.net wrote: On 08/18/10 11:07, Tor Lillqvist wrote: A motion blur is a retinal effect that has a time dependence. Is motion blur actually something people perceive with their eyes and brain, or something that only exists in physical artefacts? (Either intentionally created by an artist to give the impression of motion, or as an direct result of the method the still or motion picture was created.) And we have been so used to it that we know what it means, even if it doesn't correspond to what we actually see? (But yeah, gg's arguments make sense.) --tml Good point, the equal weighting probably is close to what a silver nitrate film camera would record, which is probably what this was intended to model. [...] The human eye has sharpness only in the middle of sight. Things that move forward and can't be focussed, obviously willo be unsharp. and translation and rotation therefore will also give unsharpness in the human eye/brain. So this is not only what a camera would record. Ciao, Oliver ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
On 18.08.2010 12:50, g...@catking.net wrote: On 08/18/10 11:07, Tor Lillqvist wrote: A motion blur is a retinal effect that has a time dependence. Is motion blur actually something people perceive with their eyes and brain, or something that only exists in physical artefacts? (Either intentionally created by an artist to give the impression of motion, or as an direct result of the method the still or motion picture was created.) And we have been so used to it that we know what it means, even if it doesn't correspond to what we actually see? (But yeah, gg's arguments make sense.) --tml Good point, the equal weighting probably is close to what a silver nitrate film camera would record I think so, too. Consider the star trails in a long-time exposure of a night sky: there is no decay visible. However, and whatever the motivation, it's an interesting idea, so here's a quick comparison for a linear motion blur: http://yahvuu.files.wordpress.com/2009/08/blurtest.png Regretably, the mathmap convolve function introduces some artifacts, but i think it can be seen that 'decaying' (or 'soft'?) motion blur is an option of artistical relevance. Tongue in cheek, i shurely wouldn't oppose if someone wanted to get code providing this functionality included into GIMP .- regards, yahvuu ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
Thanks for the insight and ideas. I didn't mention I am working on a numerical algorithm for deblurring images that have been blurred by non-accelerated circular motion around any possible axis so it was nice to get some ideas however the applications for which it is intended require more accuracy. I also noticed that this particular filter is quite slow. I have been experimenting with deforming the image into polar coordinates using interpolation schemes to get a good resolution and then applying regular, linear motion blur horizontally or vertically which would correspond to the zoom blur or rotational motion blur effects. Then the image must be deformed back to cartesian coordinates using again some sort of interpolation. So far it looks right if resolution is taken care of and it's quite fast. Haven't tried the cases where the image plane is not perpendicular to the axis of rotation and the debluring algorithm still needs to be developed. In any case all this is in prototype form (Matlab) and if all goes well it could be perhaps ported to GIMP when everything has been worked out. ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
I think so, too. Consider the star trails in a long-time exposure of a night sky: there is no decay visible. However, and whatever the motivation, it's an interesting idea, so here's a quick comparison for a linear motion blur: http://yahvuu.files.wordpress.com/2009/08/blurtest.png To complete the set, here is a non-Gaussian blur with an x blur of 40 px and a y blur of 0. http://www.majhost.com/gallery/ffaat/gimp/more2/gb.png -Rob A ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
On 18.08.2010 19:55, Rob Antonishen wrote: To complete the set, here is a non-Gaussian blur with an x blur of 40 px and a y blur of 0. thank you, new URL is http://yahvuu.files.wordpress.com/2009/08/blurtest2.png regards, yahvuu ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] Rotational Motion Blur
Hello, I would like to ask for some insight concerning the algorithm used for the Rotational Motion Blur filter if possible. I would be very grateful if someone could post the idea behind the implemenation of this filter either by a mathematical algorithm or pseudocode. Thanks in advance. Kostas ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Rotational Motion Blur
I just looked over the code -- if I understand it correctly, it creates an arc of the specified angle passing through each point of the image, generates a set of points equally spaced along the arc, uses interpolation to get the color for each of these points, and then averages all the colors together, with equal weighting. The number of points is proportional to the length of the arc, with adjustments if it is too large or too small. If you need more detail, please ask. -- Bill ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
