search

Zhaozixiang1228 / MMIF-CDDFuse

[CVPR 2023] Official implementation for "CDDFuse: Correlation-Driven Dual-Branch Feature Decomposition for Multi-Modality Image Fusion."
376 stars 37 forks source link

为什么运行测试代码的结果与原论文相差甚远呢 #48

Open tbatmmx opened 1 month ago

tbatmmx commented 1 month ago

为什么运行测试代码的结果与原论文相差甚远呢

tbatmmx commented 1 month ago

我运行出来的结果是这样的

The test result of TNO : EN SD SF MI SCD VIF Qabf SSIM CDDFuse 0.0 0.0 0.0 0.0 0.52 0.0 0.0 0.02

================================================================================ The test result of RoadScene : EN SD SF MI SCD VIF Qabf SSIM CDDFuse 0.0 0.0 0.0 0.0 0.19 0.0 0.0 0.0

以及这样 The test result of MRI_CT : EN SD SF MI SCD VIF Qabf SSIM CDDFuse_IVF 0.0 0.0 0.0 0.0 -0.9 0.0 0.0 0.87 CDDFuse_MIF 0.0 0.0 0.0 0.0 -0.9 0.0 0.0 0.87

================================================================================ The test result of MRI_PET : EN SD SF MI SCD VIF Qabf SSIM CDDFuse_IVF 0.0 0.0 0.0 0.0 -0.72 0.0 0.0 1.19 CDDFuse_MIF 0.0 0.0 0.0 0.0 -0.72 0.0 0.0 1.19

================================================================================ The test result of MRI_SPECT : EN SD SF MI SCD VIF Qabf SSIM CDDFuse_IVF 0.0 0.0 0.0 0.0 -1.31 0.0 0.0 1.13 CDDFuse_MIF 0.0 0.0 0.0 0.0 -1.31 0.0 0.0 1.13