>> "ಥ﹏ಥ.jpg"
All these glyphs in your file names are meaningless to us. We need: 1) the hex value of each byte, 2) the codepage/encoding of those bytes, 3) how those bytes were obtained/used.

If you want to know what the true Unicode filenames are (assuming NTFS is used), then run the following from the command line:
Code:
cmd /U /C dir > files.txt
The created files.txt will be UTF16LE encoded. You can then load the file in a hex editor and see the values of each WCHAR that make up the filename. Ultimately, this is what needs to be passed to CreateFileW, _wfopen, or fstream::open(const wchar_t*). Anything else is a non-solution (without renaming the files).

For the same reasons, the filenames in your URL's should be UTF8 encoded. You'll then be dealing with Unicode from beginning to end. No confusion.

gg