Blue Ridge Mtn. Wildlife

Yes, the adjustments cancelled out the distortions introduced by your conversion operations. Apparently you began with a 2D image, as that’s what resulted when auto-aligned with one click with SPM.

Regarding the remaining discussion, I have to be adamant that true, accurate 3D/stereoscopic imaging contains disparities in the horizontal plane ONLY. I consider the introduction and presence of significant (greater than say, 1%) vertical disparities as errors in stereoscopic display, to be eliminated, or rejected, for display. My reasoning (other than obvious viewing discomfort it does induce in most observers) is that we have binocular eyes, oriented strictly in a horizontal plane. So do most other high level species. I won’t touch on creatures with plenoptic eyes.

The Bonsai image has some obvious defects, as it was not taken simultaneously. For instance, the cloud shadows have some issues.

That said, there are some exceptions. One of which is a stereo still image that contains specular reflections, such as from a wavy surface of water. While retinal rivalries become common with these reflections, at times there can also be localized vertical disparities. This issue at work is it is a still image of a rapidly moving scene. Without that motion, the still image can appear both “odd”, and still be realistic, as elements surrounding these reflections have the required, and true, horizontal disparity.

Some portion of the population have defects in binocular vision (such as ambliopia, or poor ocular muscle control). They may have experienced vertical disparities due to such maladies, and have grown to accommodate their vision in ways unexpected by those with normal vision. This might be at work with any that find your wildlife conversion as realistic, or even viewable without discomfort. I have normal vision, and I find that conversion very uncomfortable to attempt to fuse, as I see the entire image to be fused at once, with no accommodation for localized distorted areas that will not fuse with the overall image.

Certainly there are operations that could be performed to your wildlife image that would produce a comfortable and realistic stereoscopic image. These operations should be based solely on inducing appropriate horizontal disparities in the stereo pair. Can you attempt that?