Best approach to food scanning? Tips for feature + turntable scans?

Hey there, I’ve been scanning using photogrammetry and Lidar scanning for some time, but I wanted something more powerful to speed up and automate the process, focusing on food dishes, where textures are essential. Have anyone of you compared the time spent to get a textured 3D scan with photogrammetry vs 3D scanner?

I’ve been scrolling through the Revopoint channel but I haven’t found videos that talk in depth how to get a good scan using the Feature mode + turntable (there are multiple ones about markers, but in my case I can’t use markers). Some questions:

-Is it better to scan the target in one go, with the turntable spinning, slowly rising the height of the scanner? Or stopping the scan then rising the scanner and resuming?
-I am using a lightbox with controlled lighting, do you recommend turning off the scanner lights?
-Is it possible to stop the turntable automatically from Revoscan on PC when pausing the scan? Or is that function not implemented yet? I’ve seen that in previous versions you could even program a routine with one full round for each angle.
-Which is the best approach for objects that have a bottom part? Getting two different point clouds and then merging? or pausing mid scan and turning the object?

Difference between photogrametry and structured light sanning is at least tenfold.

For example even with a powerfull pc when processing 240 photos for the following head the scan mesh took around 10 hours to compute

The same model scanned with the mini in 3 parts, merged and remeshed with higher quality than i could achieve with photogrametry took less than one hour from scan to mesh

I recently scanned some food with the INSPIRE, first one is texture the other is simply pointcloud color.

Food usually is easy to scan, at worst you can use cornstarch and rubbing alcohol on a spray bottle to matt up some of the surfaces of the dishes

if the food is too smooth to pick up on the feature mode you can always use the crumpled paper method to add a rich feature detail to your scan model

There are food scans in my Revopoint INSPIRE showcase: pastry, sushi, a bowl of fried rice, fruits, a root vegetable.

Unless you have several cameras or an automated turntable for shooting photogrammetry, structured light will be much faster to shoot.
For objects that have a bottom: at the moment, when merging scans using Revo Scan, you don’t get textures, only vertex colors. If textures are important for you, you should scan everything in a single session and pause mid scan.
This is what I did for this pastry (even though this specific scan isn’t perfect):

I tried scanning food with the mini, it will scan food with similar contrast, but if you want to scan for example, grapes with white choclate, the contrast is too great.
also shiney surfaces like grapes can be difficult to scan, so you may need to matte everything down with a spray.
for basic scanning of food, photogrametry app on the phone is fine.