The last time I measured ImageNet JPEGs with EXIF orientation metadata, the number of affected images was actually quite small (< 100, out of a dataset of 1.28M). There are also some duplicates, but altogether it seems fairly "clean."
The last time I measured ImageNet JPEGs with EXIF orientation metadata, the number of affected images was actually quite small (< 100, out of a dataset of 1.28M). There are also some duplicates, but altogether it seems fairly "clean."