Filename exception although file names and paths are the same or correct - windows issue

[robisen1]

I am trying to use the image_bbox_slicer in that it fits exactly my use case. My problem is that it keeps through exceptions like below no matter what filenames or paths I used. I looked at your code a little bit thinking that maybe it is my file names which are like "Scenario_10_3_1_standing_by_vehicle_150ft-0058.jpg". I did not see anything in your code that would have a problem with the file name. Do you have any suggestions on what I am doing wrong?

Exception: Each image in E:/sd_data_collection/sd_11_15_19/ripped_frames/tile_test/test_src/images must have its corresponding XML file in E:/sd_data_collection/sd_11_15_19/ripped_frames/tile_test/test_src/annotations with the same file name.

[acl21]

Are the filenames of images and their corresponding annotations similar? I mean, do you have the corresponding "Scenario_10_3_1_standing_by_vehicle_150ft-0058.xml" in the "annotations" directory.

[robisen1]

First, let me say thank you for putting this tool together. As to my issue, I believe it is Windows specific. The issue arises from how this code deals with windows paths. Here is the code:

imgs = sorted(glob.glob(img_src + '/*'))
anns = sorted(glob.glob(ann_src + '/*.xml'))

That is from helpers.py in the validate_file_names function. What happens is that on windows it adds \ instead of / and this causes the exception. Anyways I resolved it by doing:

imgs = [i.replace('\\', '/') for i in imgs]
anns = [a.replace('\\', '/') for a in imgs]

I have also been changing the code to use Pathlib and some other things. I hope to have some time next week to test on Linux/Ubuntu because I am concerned my hacks might cause issues themselves but for right now the code is doing what I need. Again thanks

[acl21]

Thank you for pointing this out. I did not test this tool on Windows and I should have mentioned that in the docs. I will add it now. But thanks again for letting me know a potential fix around the problem.

I am glad you found the tool helpful. 😊

[mikeyEcology]

I'm getting this same error even though I'm running on Linux (Ubuntu). I can check to make sure there is an xml for each image file by running this:

im_list = os.listdir(im_src)
an_list = [x.replace("xml", "JPG") for x in os.listdir(an_src)]
list(set(im_list) - set(an_list))

Which returns []. I think this means that I don't have any images without a corresponding xml file?
But when I run

slicer.config_dirs(img_src=im_src, ann_src=an_src,
                   img_dst=im_dst, ann_dst=an_dst)

I get

in validate_file_names
raise Exception(
Exception: Each image in .../images/ must have its corresponding XML file
in .../labels/ with the same file name.

[mikeyEcology]

I solved my problem, so please ignore unless you're having a similar problem and want to try what I did. The slicer function is making sure that annotations and images filenames are exactly the same (with the exception of jpg and xml). If you're getting a similar error, you can try running this:

im_list = os.listdir(im_src)
an_list = [x.replace("xml", "JPG") for x in os.listdir(an_src)]
list(set(im_list) - set(an_list))
list(set(an_list) - set(im_list))

I had a couple of annotation files that python doesn't seem to think match up with image files (they do in terminal, but not in python), so I just deleted these two files, and it is working now.