A pruned YOLOv12n with augmentation and self-adversarial training reports 94% mAP@0.5 and 4.67 MB size on the DroneTrashNet desert-litter dataset, claiming real-time deployment.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
YOLO-SAT: A Data-based and Model-based Enhanced YOLOv12 Model for Desert Waste Detection and Classification
A pruned YOLOv12n with augmentation and self-adversarial training reports 94% mAP@0.5 and 4.67 MB size on the DroneTrashNet desert-litter dataset, claiming real-time deployment.