AFMPM: adaptive feature map pruning method based on feature distillation

Guo, Yufeng; Zhang, Weiwei; Wang, Junhuang; Ji, Ming; Zhen, Chenghui; Guo, Zhengzheng

doi:10.1007/s13042-023-01926-2

Cited by 1 publication

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Feature distillation, a form of knowledge distillation, involves transferring intermediate-layer features from the teacher model to the student network for learning. This enables the student model to directly align with intermediate feature map information and acquire the feature extraction capabilities of the teacher network (Guo et al, Guo YF, Zhang WW, Wang JH et al (2024) AFMPM: adaptive feature map pruning method based on feature distillation. International Journal of Machine Learning and Cybernetics 15:573-588. https://doi.org/10.1007/s13042-023-01926-2).…”

Section: Introductionmentioning

confidence: 99%

A lightweight fine-grained recognition algorithm based on object detection

Ren,

Shi,

Chen

et al. 2024

Preprint

View full text Add to dashboard Cite

In order to enhance the fine-grained recognition of fish species, this paper proposes a lightweight object detection model YOLOv8n-DFG. The model accurately identifies six deep-sea fish species including Flatfin sailfish, Striped marlin, Atlantic bluefin tuna, Bigeye tuna, Longtail tuna, and Yellowfin tuna, meeting the requirements for real-time detection and fishing monitoring. Firstly, by introducing FasterNetBlock and EMA attention mechanism into the YOLOv8 network structure to improve C2f and obtain the C2f-FE module, this model enhances feature extraction accuracy and operational efficiency. Subsequently, it combines BiFPN structure with C2f-FE module to construct a fast and lightweight neck network structure that achieves multi-scale feature fusion. Additionally, Dysample dynamic upsampling module is introduced along with porting of Adown downsampling module from YOLOv9 to optimize feature pyramid sampling method named as YOLOv8-FG. Finally using large-sized YOLOv8s-FG as teacher network and small-sized YOLOv8n-FG as student network based on CWD loss intermediate layer feature distillation method constructs the final model YOLOv8n-DFG. Experimental results on a dataset containing six morphologically similar fish species demonstrate the effectiveness of these improvements and distillation effects are significant. Compared to YOLOv8n, precision has increased by 7.8%, recall by 3.3%, mAP@50 by 5.6%, while FlOPs decreased by 42% with a reduction in model size of 58%. The results indicate that our proposed YOLOv8n-DFG demonstrates exceptional accuracy and real-time performance, effectively fulfilling the requirements for real-time fine-grained fish recognition.

show abstract

Section: Introductionmentioning

confidence: 99%