4 ms·
There's a small gallery of success and failure cases here [1]. It certainly doesn't look as good as the original, yet. I wonder if that's due to the implementa
by wongarsu 4y ago
There's a small gallery of success and failure cases here [1].
It certainly doesn't look as good as the original, yet. I wonder if that's due to the implementation differences noted, less cherry picking in what they show, or inherent differences between Imagen and Stable Diffusion.
Maybe Imagen just has a much better grasp of how images translate to actual objects, where Stable Diffusion is stuck more on the 2d image plane.
1: https://github.com/ashawkey/stable-dreamfusion/issues/1 https://github.com/ashawkey/stable-dreamfusion/issues/1
- nopenopenopeno 4y agoOne cannot help but notice the success cases are expected to have symmetry along at least one axis, whereas the failure cases are not.
- namarie 4y agoAren't squirrels and frogs expected to have an axis of symmetry? I think the reason for the failures is the presence of faces; it seems to be trying to make a face visible from all angles.
- gpm 4y agoWhich probably has a lot to do with us taking nearly all our pictures of things with faces from the face facing direction.
- shadowgovt 4y agoI feel like the Cthulhu head is extra-successful, given the subject matter. Non-Euclidean back-polygon imaging? Good work, algorithm. ;)
- codeflo 4y agoI guess a lot can be done to force the model to create properly connected 3D shapes instead of these thin protruding 2D slices. But I noticed something else. Some of the angles “in between” the frog faces have three eyes. I wonder if part of the issue might be that those don’t look especially wrong to Stable Diffusion. It’s often surprisingly confused about the number of limbs it should generate.