Open Acly opened 3 months ago
Would be great
Did some tests with exported ONNX models. They're working fine, but performance wise it would be a big regression.
Model | Size | Inference time |
---|---|---|
MobileSAM | 43 MB | 450 ms |
SAM2 Small | 179 MB | 1240 ms |
SAM2 Tiny | 151 MB | 1000 ms |
This is on CPU, which is most common scenario. The masks are mostly the same regarding pixel quality. The detection accuracy is expected to better for SAM2.
A distilled/quantized version of SAM2 model would be interesting...
Thank you, I 'll try it
Note: Meta has released updated models (SAM 2.1)
https://github.com/facebookresearch/segment-anything-2
So for precise masks some kind of follow up would still be needed.
No ONNX model out yet.