Thank you very much for your excellent work. I encountered an issue during the model training phase: the reconstruction of non-reference views in the training results is relatively reliable, but the reconstruction of the reference view is not fully aligned with that of other views. Have you experienced similar problems in your work? I look forward to your reply
Thank you very much for your excellent work. I encountered an issue during the model training phase: the reconstruction of non-reference views in the training results is relatively reliable, but the reconstruction of the reference view is not fully aligned with that of other views. Have you experienced similar problems in your work? I look forward to your reply