3D photographic inpainting from a single source image

Re:3D photographic inpainting from a single source image by yorwba on Saturday April 11, 2020 @05:49AM (#22840727)

> So basically, if you have impossible to get input data, this network can do its magic.

Except they evidently got the input data for the examples in the paper, so it can't be impossible to get.

They cite at least two different methods for adding depth information to a single image to generate the necessary RGBD data, different views of which can then be rendered with their inpainting applied:

- https://arxiv.org/abs/1907.01341

- https://research.cs.cornell.edu/megadepth/

[ Parent | on HN ]

Re:3D photographic inpainting from a single source image by fxtentacle on Sunday April 12, 2020 @07:08AM (#22847404)

Yes, you can always create that data by hand, but it's too expensive and, hence, impossible to scale.

As for using other AIs, they tend to not work too well on more complex images.

But in any case, getting the data that you need to be able to use the paper here is very challenging.

Edit: I should probably say that I have hands-on experience with MegaDepth and Midas and that it was underwhelming. Both of them assume a Gradient from close to far from the bottom to the top and both of them assume that optical variation will be in the foreground. A photo of a dining table from the side is already enough to confuse both of them.

[ Parent | on HN ]