Images with depth maps

General
1

Just a random thought here… It’s possible on many modern phones to take photos with depth maps. How these are saved varies by manufacturer, i think, but in a world with unlimited resources, it would be neat if the iNat app could handle this sort of thing in a consistent implementation across hardware / OS. I assume this would require iNat to allow saving images on the platform in a format that supports this, since i don’t think base JPEG has such support.

But I guess the harder part is to display such photos in an intuitive way. Facebook’s 3D Photos is the only major implementation of a way to visualize these photos that I’ve heard of. But it would be neat for iNat to implement something (maybe something as simple as a toggle between standard image and depth visualization), that works in both mobile and desktop, since it might allow folks to visualize the subject of observations a little better. (I guess in the meantime, the only way to provide that sort of depth information is to take a couple of photos that are slightly offset and post both in the same observation.)

1 Like
2

Depth maps, like a spatial view?

Personally, I vote for having videos on the platform first bc it’s multiple steps to turn a video into a gif, and then not have it be so large in MB size that it can’t upload.

File size would be a concern with depth maps I would think also…

3

it’s like a layer that depicts depth.

it looks like Google adds depth information via XMP metadata in JPEG files. unfortunately, that information is stripped by exiftool when it is used to strip metadata, unless XMP is explicitly preserved. so maybe JPEG can be used as long as XMP depth information isn’t stripped upon upload.

…

i’m not against videos in concept, but that’s best discussed in one of the existing forum topics that covers that, not here.

in my mind, video is a whole different beast. depth in images isn’t that far off from what iNat currently does at its core, and that additional information could be used for all sorts of things potentially.

i assume that iNaturalist at some point in the future will incorporate depth into their app’s AR camera. i imagine a view when making observations that highlights multiple subjects at once and identifies them all with labels pinned to the subjects in 3D way. so capturing depth when taking images is in the same path that leads to that sort of future.

1 Like
4

https://github.com/inaturalist/inaturalist/issues/2858

1 Like
5

that format is supported now, but

Note that these images will be converted to JPEGs, they don’t remain HEIC/HEIF files on iNat’s servers.

4 Likes
6

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.