light probe of objects on a table

Hi there

I'm trying to create a light probe similar to Debevec's Rendering synthetic
objects into real scenes and using the corresponding points in HDR shop I
create the resulting probe following instructions on
http://www.ict.usc.edu/graphics/HDRShop/tutorial/tutorial5.html

Taking a closeup orthographic projection photo of the probe and then another at
90 degrees rotated about the probe I match points and blend images using the
formula with the given image mask. In the result I get 2 of everything, objects
& light sources. I must be doing something wrong. Is it something to do with
selecting the right points, or the positioning of the probe?

Thanks for any help

Tarik

···

--
Tarik Rahman
PhD student, Institute of Perception, Action and Behaviour
School of Informatics
University of Edinburgh

Tarik,

You will likely have to create your own mask to clone out the features you want to eliminate (as opposed to using the one in the tutorial). Creating the mask is easily done in Photoshop or similar app that will permit you to create an 8bit grayscale image where black is no pass and white is full pass.

Take your HDRs and convert them to LDRs and bring them into Photoshop, etc. as reference for creating your mask.

I assume that you kept the camera at the same elevation and simply moved it 90 degrees from its original position, orbiting in the horizontal (ground) plane. Then, it should just be a matter of marking the reference points systematically and consistently. If you want, send me your two probe images and I'll take a look.

Good luck,

kirk

···

On May 23, 2005, at 10:44 AM, Tarik Rahman wrote:

Hi there

I'm trying to create a light probe similar to Debevec's Rendering synthetic
objects into real scenes and using the corresponding points in HDR shop I
create the resulting probe following instructions on
http://www.ict.usc.edu/graphics/HDRShop/tutorial/tutorial5.html

Taking a closeup orthographic projection photo of the probe and then another at
90 degrees rotated about the probe I match points and blend images using the
formula with the given image mask. In the result I get 2 of everything, objects
& light sources. I must be doing something wrong. Is it something to do with
selecting the right points, or the positioning of the probe?

Thanks for any help

Tarik
--
Tarik Rahman
PhD student, Institute of Perception, Action and Behaviour
School of Informatics
University of Edinburgh

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

[email protected]

···

On May 23, 2005, at 8:07 PM, Kirk Thibault wrote:

Tarik,

You will likely have to create your own mask to clone out the features you want to eliminate (as opposed to using the one in the tutorial). Creating the mask is easily done in Photoshop or similar app that will permit you to create an 8bit grayscale image where black is no pass and white is full pass.

Take your HDRs and convert them to LDRs and bring them into Photoshop, etc. as reference for creating your mask.

I assume that you kept the camera at the same elevation and simply moved it 90 degrees from its original position, orbiting in the horizontal (ground) plane. Then, it should just be a matter of marking the reference points systematically and consistently. If you want, send me your two probe images and I'll take a look.

Good luck,

kirk

On May 23, 2005, at 10:44 AM, Tarik Rahman wrote: