🧪 Test?View on arXiv
Cross-Generation Optimization of YOLOv26, YOLOv11, and YOLOv8 for Fine-Grained Small-Object Detection and Instance Segmentation in Complex Orchards
rnjnspkt
fine-tuningobject detectioninstance segmentation
2608.23636
Builder Relevance
3h ago80%
Abstract
This study benchmarks YOLO models for detecting and segmenting small objects in orchard environments.
Reality Card
Core Claim
YOLOv11s-960 achieved the highest mask mAP@50:95 of 0.402 and box mAP@50:95 of 0.426, demonstrating effective small-object detection.
Method / Result
YOLOv11s-960 achieved mask mAP@50:95 of 0.402 and box mAP@50:95 of 0.426.
Limitations
Increasing model capacity did not consistently improve accuracy, indicating potential issues with model scaling.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.