Tweet
I have recently joined a forum that specializes in macro photography. I heard about a photo stacking software called Helicon Focus that can output a 3D file from one stack of photos. Using the areas of the frame that are in sharp focus it is able to build the 3D information using a Depth map stacking process. I can not help but notice that part of 3DF Zephyr also processes something called Depth Map. I wonder if these processes are at all similar, and if the Helicon Focus output could some how be used to build complete 3D information. It has an obvious limitation, in that is only able to process information from one direction, but the information seems like it would have some very accurate topographical information as the different narrow layers are processed.that are in sharp focus. I have seen hand held 3D scanners that record information from various directions of the subject, and are then able to reconstruct a complete representation by stitching together the multiple scanner passes. I wonder if there is expertise here that could address this technology, and if in the future this might be a possible solution for building 3D macro surfaces. I would envision that this technology could make it possible to describe the object with much fewer camera positions.
-
-
Hi Carl,
from a quick look, it simply seems that it's a software for focus stacking, which is a technique that combines more images at different focal lenght to create a new image with a better depth of field.
While we do have some example datasets created with focus stacked images, you have to be careful - most dataset I had a look at had distortions not due to the lense (which zephyr can and will fix - in fact it's a needed information for zephyr) but due to the software that created the stacked image.
I personally would simply feed the images you take at different focal lenght, if it is not possible to take one clear, sharp image and eventually consider masking parts of the image that are very out of focus, as it may bring noise during texturing (although zephyr is robust to noise and blur, sometimes it may be necessary in extreme cases) -
Helicon Focus has a unique ability, in addition to focus stacking photos, it can generate a 3D file from the data from only a single perspective.
There is very likely distortion in this process, as either your camera has to move our your have to shift your focus ring in tiny increments to capture macro subjects. There is no way that I have heard of to capture more than a sliver of a higher magnification subject without stacking. Some of my final images took 70 photos to achieve sharp focus in the entire picture, and my magnification level is still at an amateur level. The pros use microscope optics, and computer controlled step mechanisms to capture incredible 2D images. These precision steps can output height information because only an incredibly narrow band of the subject is in focus during any one exposure. I think this process could have a lot of potential to build models of high magnification subjects. I just thought someone at your company may want to consider this approach, and discuss it further.
This is a web page that discusses this approach. http://www.gigamacro.com/blog/gigapi...rt-case-study/Comment
Comment