Performances of different loss functions.

<div><p>As object detection models grow in complexity, balancing computational efficiency and feature expressiveness becomes a critical challenge. To address this, we propose SAMF-YOLO, a novel model integrating three key components: SONet, BFAM, and FASFF-Head. The UniRepLKNet backbone,...

Täydet tiedot

Tallennettuna:

Bibliografiset tiedot
Päätekijä:	Jun Huang (127297) (author)
Muut tekijät:	Shamsul Arrieya Ariffin (21637732) (author), Qiang Zhu (188733) (author), Wanting Xu (3180711) (author), Qun Yang (415944) (author)
Julkaistu:	2025
Aiheet:	Biochemistry Neuroscience Sociology Space Science Biological Sciences not elsewhere classified Information Systems not elsewhere classified experimental results demonstrate challenging object scales scale feature fusion feature expressiveness becomes yolo surpasses yolov11s div >< p balancing computational efficiency feature space computational cost unireplknet backbone star operation small defects precision approach notable reduction minimal overhead labeled data critical challenge
Tagit:	Lisää tagi Ei tageja, Lisää ensimmäinen tagi!

Performances of different loss functions.

Samankaltaisia teoksia