模型压缩用于实时对象检测,使用严格的分级修剪
Defu Yang1, Mahmud Iwan Solihin1, Yawen Zhao1
1Faculty of Engineering, Technology and Built Environment, UCSI University, Kuala Lumpur, Malaysia.
iScience
|January 21, 2025
概括
严格分级修剪 (RGP) 显著压缩了像YOLOv8这样的对象检测模型,实现了高精度和更快的处理速度. 这种方法有效地削减了冗余过器,以实现高效的实时检测.
科学领域:
- 计算机视觉 计算机视觉
- 机器学习 机器学习
- 深度学习 (Deep Learning) 是一种深度学习.
背景情况:
- 实时对象检测需要平衡模型大小和准确性.
- 深度学习模型中的卷积层通常具有不均的贡献和低冗余性,使压缩工作复杂化.
研究的目的:
- 引入严格分级修剪 (RGP) 作为压缩对象检测模型的有效方法.
- 为了保持高检测准确度,同时显著减少模型大小和增加处理速度.
主要方法:
- RGP使用非敏感的第一阶泰勒近似来评估精确的内核修剪的过器重要性.
- 该方法反复评估层的意义,以保护关键层并确保强大的检测性能.
- RGP应用于YOLOv8对象探测器,并在GTSDB,Seaships和COCO数据集上进行验证.
主要成果:
- 在GTSDB数据集上,RGP实现了YOLOv8n的80%压缩,最小的0.11% mAP0.5下降,FPS增加了43.84%.
- 对于YOLOv8x,RGP实现了90%的压缩,mAP0.5:0.95增加了1.26%,FPS提升了112.66%.
- 在海船和COCO数据集上也观察到相当大的压缩率,证实了RGP的多功能性.
结论:
- 严格分级修剪是一种强大的技术,用于在各种数据集中压缩对象检测模型.
- 在不影响准确性的情况下,RGP显示了开发高效,高速的物体检测系统的巨大潜力.
相关概念视频
Classification of Systems-II
133
Continuous-time systems have continuous input and output signals, with time measured continuously. These systems are generally defined by differential or algebraic equations. For instance, in an RC circuit, the relationship between input and output voltage is expressed through a differential equation derived from Ohm's law and the capacitor relation,
133
Force Classification
1.1K
Forces play a crucial role in the study of physics and engineering. They are essential in describing the motion, behavior, and equilibrium of objects in the physical world. Forces can be classified based on their origin, type, and direction of action.
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
1.1K
Difference from Background: Limit of Detection
5.8K
The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
The LOD indicates the presence or absence...
The LOD indicates the presence or absence...
5.8K
Reducing Line Loss
143
In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
143
Classification of Systems-I
168
Linearity is a system property characterized by a direct input-output relationship, combining homogeneity and additivity.
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:
168
Design Example: Aggregate Gradation
90
The right type and quality of aggregates are crucial for concrete as they significantly influence its properties, mix proportions, and cost-effectiveness. If different sources are available for sand, the commonly used fine aggregate in concrete, the selection of sand is primarily based on its gradation.
The grading, or particle-size distribution, of sand is determined using sieve analysis, with standard sizes ranging from 150 μm to 10 mm (ASTM No. 100 sieve to 3⁄8 in. sieve). Sand is...
The grading, or particle-size distribution, of sand is determined using sieve analysis, with standard sizes ranging from 150 μm to 10 mm (ASTM No. 100 sieve to 3⁄8 in. sieve). Sand is...
90


