MFF多焦点图像融合遇到的问题

[baby1634810490]

您好，非常感谢您的指点，有些问题想请教一下，就是关于MFF多焦点融合，您的github项目里面写的是需要把训练数据集放入MEF中训练，但是trainsets中是有MFI-WHU的。我把数据集放入了MFI-WHU中训练的，成功出了模型，但是在测试过程中出现了数组越界的错误（IndexError: list index out of range），我想请问一下，是不是需要放入MEF中进行训练出的模型才可以测试成功呀，不能把数据放入MEI-WHU中训练呀。还是我缺少了些步骤呀。我的错误具体如下： [1/2] Saving fused image to : results/SwinFusion_Lytro/1-checkpoint.jpg, Processing time is 557.452528 s Traceback (most recent call last): File "/root/horse/SwinFusion/SwinFusion-master/test_swinfusion.py", line 158, in main() File "/root/horse/SwinFusion/SwinFusion-master/test_swinfusion.py", line 69, in main for i, test_data in enumerate(test_loader): File "/opt/conda/envs/pytorch1.8/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 517, in next data = self._next_data() File "/opt/conda/envs/pytorch1.8/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1199, in _next_data return self._process_data(data) File "/opt/conda/envs/pytorch1.8/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1225, in _process_data data.reraise() File "/opt/conda/envs/pytorch1.8/lib/python3.9/site-packages/torch/_utils.py", line 429, in reraise raise self.exc_type(msg) IndexError: Caught IndexError in DataLoader worker process 0. Original Traceback (most recent call last): File "/opt/conda/envs/pytorch1.8/lib/python3.9/site-packages/torch/utils/data/_utils/worker.py", line 202, in _worker_loop data = fetcher.fetch(index) File "/opt/conda/envs/pytorch1.8/lib/python3.9/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] File "/opt/conda/envs/pytorch1.8/lib/python3.9/site-packages/torch/utils/data/_utils/fetch.py", line 44, in data = [self.dataset[idx] for idx in possibly_batched_index] File "/root/horse/SwinFusion/SwinFusion-master/data/dataloder.py", line 37, in getitem B_path = self.paths_B[index] IndexError: list index out of range

非常感谢您的指点，麻烦您了，谢谢您。

[Linfeng-Tang]

你这是训练还是测试过程中出现的错误呢？ 你可以在dataloader中打印self.paths_B 和self.path_A 看看二者是否是一样的

[baby1634810490]

您好，非常感谢您的指点，这个错误是才MFF测试中遇到的，打印出的两个路径也不是完全一样的（self.paths_A is -> ['Dataset/testsets/Lytro/A_Y/.ipynb_checkpoints/1-checkpoint.jpg', 'Dataset/testsets/Lytro/A_Y/1.jpg'] self.paths_B is -> ['Dataset/testsets/Lytro/B_Y/1.jpg']）请问这个问题是什么原因造成的呢？

[Linfeng-Tang]

这个是由于 你A_Y数据集里多了一个'Dataset/testsets/Lytro/A_Y/.ipynb_checkpoints/1-checkpoint.jpg', 造成的 你可以尝试把.ipynb_checkpoints 文件夹删除来解决这个问题呢

[baby1634810490]

   谢谢您的指点，我那个里面数据集没有这个文件（A_Y数据集里多了一个'Dataset/testsets/Lytro/A_Y/.ipynb_checkpoints/1-checkpoint.jpg',），目前我在datalodar里面给B_path的数组索引进行-1操作，测试程序完美执行了，我还想问问，输出的两个图片（0.png和1.png）为灰度图像，是不是也需要进行YCBCR转换成RGB呀，而且测试代码为啥会保存两张图像呀，融合后不应该是一张图像吗？
   非常谢谢您的指点

[baby1634810490]

非常感谢您的指点，我已经解决了YCbCr的问题了，出来的效果也不错，再次感谢您的指点

[Linfeng-Tang]

好的 好的