pikaliov 3b5778e303 Unify segmentation classes (17) across GTA-UAV and UAV_VisLoc ...

Extract shared UNIFIED_PROMPTS (17 classes, ID 0-16) into
scripts/seg_classes.py for transfer learning compatibility.
Both run_gta_uav.py and run_uav_visloc.py now import from it.

Key change: swimming pool moved from ID 13 → ID 16, so sports field
(ID 13), muddy ground (14), embankment (15) have stable IDs across
both datasets. Missing classes in a dataset = 0 pixels = 0 loss.

Updated: README, CLAUDE.md, segmentation_class_analysis.md, palette.
Deleted old UAV_VisLoc segmentations (need regeneration with 17 classes).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-04-17 22:06:10 +03:00

1.1 KiB

Raw Blame History

View Raw