I was wondering how people feel about nano banana’s ability to reference a face from an attached image and use that for its image generation?
I must say it’s close but I often end up with a more generic looking face than the one referenced.
Wanted to see if anyone out there has any tips on having nano banana’s face referencing.
Do you submit multiple references from different angles?
Is it all in the prompting?
Would love to hear what everyone thinks.
Thanks!