'Image segementation dataset format
I'm following this tutorial: Tensorflow Image Segmentation
I want to make my own dataset. Ideally, following the same format as the Oxford pet dataset used in the tutorial in order to avoid conflicts. The problem is, I'm not sure what format is that, if anyone could tell me it would make research much easier.
From what I've been reading there is the COCO format and it seems matlab has a format of their own, but I'm still pretty confused. I'm beggining to think that maybe segmentation datasets don't always follow a standarized format, is that the case?
If so, what tools do you recommend for making a dataset like that? From what I've seen it uses trimaps and xml for bounding boxes. Besides a txt file listing each file and its labels.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|