Have you tried training with mobile-sam?

Question

Have you tried training with mobile-sam?

sijie-Xu opened this issue 7 months ago · comments

Have you tried using mobile-sam in Ultralytics? Does it support access to low-level features? I've noticed that the training process is quite lengthy with a large dataset. Currently, I'm unsure if mobile-sam can improve training efficiency compared to fastsam.

Lei · Answer 1 · Wed Jan 10 2024 15:04:57 GMT+0800 (China Standard Time)

Hi. I have not tried this yet. I may try mobileSAM and efficientSAM and see if there are possible improvements.
If you are interested, you can also try and share the results.

Lei · Answer 2 · Tue Apr 30 2024 11:42:48 GMT+0800 (China Standard Time)

Have you tried using mobile-sam in Ultralytics? Does it support access to low-level features? I've noticed that the training process is quite lengthy with a large dataset. Currently, I'm unsure if mobile-sam can improve training efficiency compared to fastsam.

I have added support to efficientSAM. You are welcome to check: https://github.com/ggsDing/SAM-CD/blob/main/models/effSAM_CD.py