When uploading image files to a photogrammetry asset, each of your image files need to be classified as either a photo or a texture. This ensures that the images are processed and displayed correctly within Nira's UI. Both of Nira's classification methods (Automatic and Manual) are documented below.
Automatic image file classification
By default, Nira will attempt to automatically classify each of your image files as either a photograph or a texture map. Nira makes this determination using the rules below. The rules are processed in order, and whichever rule matches first "wins".
For each image file,
- If the image filename is present in a camera CSV file that you're also uploading to the asset, it's a photo.
- If the image filename is present in an mtl file that you're also uploading to the asset, it's a texture.
- If the image filename starts with "DJI_", it's a photo.
- If the image width and height are the same and are a power-of-two, it's a texture. Some example dimensions that qualify are 1024x1024, 4096x4096, or 8192x8192.
- If an image has a 'Camera Model' EXIF tag present, it's a photo.
- If the image width and height are both power-of-two, it's a texture. Some example dimensions that qualify are 4096x8192 or 1024x2048.
- If the image width and height are the same, it's a texture. Some examples are 1234x1234 or 2011x2011.
These rules work quite well in most cases, but if you try to upload images that don't match any of the rules above, you'll receive the following message: "Some images could not be classified as photos or textures." as shown in the screenshot below. To resolve this, carefully follow the instructions given in the dialog to classify any listed image files.
Manually classifying your image files
In rare circumstances, the rules above might classify one of your image files incorrectly. For example, you may have an texture file with a 'Camera Model' EXIF tag on it and dimensions of 2048x1024, and you haven't provided an mtl file. This image would match #5 above and be classified as a photo.
If this occurs, you can manually classify all of your image files rather than using the default automatic classification. To do this, after dragging your files into Nira, open the file options, check the "Manually classify image files" box, then click Next, as shown in the screenshots below. On the next page, follow the instructions carefully to classify your image files.