Directory name collision between YOEO and TORSO-21

[Jan Gutsche]

• YOEO expects segmentation mask images inside a directory called segmentations (see here).
  • Mask are machine-seeable: very dark colors (0, 1, 2, ...)
  • createYoeoLabelsTORSO.py script (before #60) outputs them at yoeo_masks directory (see here).
• TORSO-21 provides segmentation mask images inside a directory called segmentaions
  • Masks are human-seeable: colors are distinct (0, 127, 255, ...)

Expected behavior

The directory names should not collide, i.e., should not be identical.

Current behavior

Converted masks are in yoeo_masks directory.
User has to rename directory and resolve name conflict manually.

Possible Solution

• YOEO's dataloader reads from a differently called directory like yoeo_masks
• ...

[Florian Vahl]

Honestly, I think this resulted in YOEO using the original TORSO annotations...
See line https://github.com/bit-bots/YOEO/blob/main/yoeo/utils/datasets.py#L125

[Florian Vahl]

But I think that YOEO should not use the TORSO annotations directly.

[Jan Gutsche]

In person conversation with @Flova:

• The convert script should put it's output (0,1,2, ...) to yoeo_segmentations, where the dataloader will be reading from.
• We should not map the TORSO-21 segmentations (0, 127, 255, ...) to YOEO format inside the dataloader (as mentioned in #61 (comment)). By this, YOEO directly depends on TORSO-21, which should not be the case, and we have implemented the conversion in two times, circumventing the script.

[Jan Gutsche]

Partially resolved by: #60