LoadImageArea?

For the most part you have to load all of the image data in order to decode it. To load a row of pixels in a PNG file for example, if there is filtering and compression present, you have to decode all of the pixels in every row in order to calculate the decompression. Likwise with JPEG, you have to read chunks in order to decompress. The only thing you'd be able to do any different is only output the parts you want to a buffer rather than all of it, but either way you still have to load the whole file and decompress all of it. The only exception to that is to use a file format that doesn't have any compression that relies on any information from adjacent pixels, which is pretty impossible, ie like a RAW file. You could load a window within RAW with no problem.