Hi, thanks for open-source such wonderful code base!
I was wondering why the renderings are quite different when I set the rendering resolution differently?
Specifically, when I train at 1k resolution, and render on 1k(see the left figure), eveything looks fine. However, when I render at 4k resolution (still use the 1k checkpoint), the results turns much darker (see the right figure).
From my viewpoint, mipnerf/zipnerf can be trained with different resolutions. So the rendering at different resolutions should be also fine?
Hi, thanks for open-source such wonderful code base! I was wondering why the renderings are quite different when I set the rendering resolution differently? Specifically, when I train at 1k resolution, and render on 1k(see the left figure), eveything looks fine. However, when I render at 4k resolution (still use the 1k checkpoint), the results turns much darker (see the right figure).
From my viewpoint, mipnerf/zipnerf can be trained with different resolutions. So the rendering at different resolutions should be also fine?
Do you have any idea why this happens? Thanks.