多层特征级联融合尖端神经网络用于对象检测
Yongqiang Ma1, Bailin Guo1, Xuetao Zhang1
1State Key Laboratory of Human-Machine Hybrid Augmented Intelligence, National Engineering Research Center of Visual Information and Applications, Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University, Xi'an, Shaanxi 710049, P. R. China.
International journal of neural systems
|September 18, 2025
概括
这项研究引入了一种用于对象检测的新型尖端神经网络 (SNN),用级联操作取代非尖端残余连接. 新模型通过确保纯粹的基于尖峰的计算和改善梯度流程来实现最先进的结果.
科学领域:
- 计算机科学 计算机科学
- 人工智能的人工智能
- 计算神经科学是一种神经科学.
背景情况:
- 尖端神经网络 (SNN) 是生物启发的模型,以低功耗,事件驱动的计算而闻名.
- 传统的对象检测网络经常使用残余结构,引入了挑战SNN实施的非尖端操作.
- 将SNN集成到对象检测中需要克服残余连接与纯粹的基于尖端处理的不兼容性.
研究的目的:
- 开发一个用于对象检测的尖端神经网络架构,可以消除非尖端操作.
- 为了提高深度SNNs中的梯度传播和特征保存,以提高检测准确性.
- 提出一种新的SNN模型,在整个网络中保持纯粹的尖端计算.
主要方法:
- 引入了多级级级联的特征提取模块,以取代级操作的剩余连接.
- 开发了一个聚合-卷积模块,将最大聚合和尖端卷积结合起来,以实现有效的下方采样.
- 通过重新设计特征提取和降低样本流程,确保纯粹的基于尖峰的计算.
主要成果:
- 拟议的多层特征级联融合SNN (MFCF-SNN) 在对象检测任务上展示了最先进的性能.
- 通过级联运算消除非尖端计算,增强了梯度传播.
- 聚合-卷积模块有效地保留了特征信息,并改善了深度SNN中的梯度流.
结论:
- MFCF-SNN有效地推进了基于SNN的对象检测,通过纯粹的尖端计算来实现深度网络训练.
- 新型模块成功地解决了SNN传统网络中残余结构的挑战.
- 该方法验证了SNNs在高性能,低功耗物体检测应用中的潜力.
相关概念视频
01:22Force Classification
2.3K
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,...
2.3K
01:21Association Areas of the Cortex
8.9K
Association areas are regions of the cerebral cortex that do not have a specific sensory or motor function. Instead, they integrate and interpret information from various sources to enable higher cognitive processes such as memory, learning, and decision-making. Some key association areas include the following:
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
8.9K