Hello, I am currently trying to fine-tune Cellpose-SAM on the PlantSeg dataset in order to reproduce the results reported in your paper.
At the moment, I am focusing on reproducing the 2D segmentation performance. Following the method described in the paper, I extracted 2D slices from the volumetric data. However, I am not sure whether I extracted the data in exactly the same way as used in your experiments.
I also tried fine-tuning the model using your released code, but the results I obtained are still significantly worse than those reported in the paper.
Could you please provide more details about the data preprocessing or slice extraction procedure used for the 2D experiments? Any advice or clarification would be greatly appreciated.
Thank you very much for your time and help.
Hello, I am currently trying to fine-tune Cellpose-SAM on the PlantSeg dataset in order to reproduce the results reported in your paper.
At the moment, I am focusing on reproducing the 2D segmentation performance. Following the method described in the paper, I extracted 2D slices from the volumetric data. However, I am not sure whether I extracted the data in exactly the same way as used in your experiments.
I also tried fine-tuning the model using your released code, but the results I obtained are still significantly worse than those reported in the paper.
Could you please provide more details about the data preprocessing or slice extraction procedure used for the 2D experiments? Any advice or clarification would be greatly appreciated.
Thank you very much for your time and help.