Does it make sense to convert the image to grayscale?

[FelipeErmeson]

Apparently the convnext_xlarge_in22k model, for example, was trained on colored images, so it would be best to leave the image colored. I also don't see any operator that transforms the image to grayscale.

[junjiejiangjjj]

Image conversion can be done using cv2, defining a custom operator, and using it in the pipeline.(https://towhee.readthedocs.io/en/latest/operator/usage.html#custom-operators)