A compression pipeline for one-stage object detection model

Li, Zhishan; Sun, Yi-Ran; Tian, Guanzhong; Xie, Lei; Liu, Yong; Su, Hongye; He, Yifan

doi:10.1007/s11554-020-01053-z

Cited by 11 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The abovementioned approaches prune CNN models from different perspectives, laying the foundation for researchers to prune more complex object detection models. The papers [5,8,34] evaluated the importance of filters in YOLOv3 using the scale factors of BN layers. Then, they remove the filters with low importance and use various optimization to recover the accuracy of models.…”

Section: Model Pruningmentioning

confidence: 99%

CAP-YOLO: Channel Attention Based Pruning YOLO for Coal Mine Real-Time Intelligent Monitoring

Meng

et al. 2022

Sensors

View full text Add to dashboard Cite

Real-time coal mine intelligent monitoring for pedestrian identifying and positioning is an important means to ensure safety in production. Traditional object detection models based on neural networks require significant computational and storage resources, which results in difficulty of deploying models on edge devices for real-time intelligent monitoring. To address the above problems, CAP-YOLO (Channel Attention based Pruning YOLO) and AEPSM (adaptive image enhancement parameter selection module) are proposed in this paper to achieve real-time intelligent analysis for coal mine surveillance videos. Firstly, DCAM (Deep Channel Attention Module) is proposed to evaluate the importance level of channels in YOLOv3. Secondly, the filters corresponding to the low importance channels are pruned to generate CAP-YOLO, which recovers the accuracy through fine-tuning. Finally, considering the lighting environments are varied in different coal mine fields, AEPSM is proposed to select parameters for CLAHE (Contrast Limited Adaptive Histogram Equalization) under different fields. Experiment results show that the weight size of CAP-YOLO is 8.3× smaller than YOLOv3, but only 7% lower than mAP, and the inference speed of CAP-YOLO is three times faster than that of YOLOv3. On NVIDIA Jetson TX2, CAP-YOLO realizes 31 FPS inference speed.

show abstract

Section: Model Pruningmentioning

confidence: 99%