Unverified50% confidenceFactExact time
CNN的核心创新在于权重共享与局部连接机制，通过滑动卷积核在图像不同区域共享同一组参数
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
UnverifiedCNN通过局部感受野和权重共享机制高效提取图像的空间层级特征，是YOLO等检测模型的早期基础76% similarUnverifiedConvolutional Neural Networks (CNNs), inspired by the structure of the visual cortex, extract local features by sliding convolutional kernels across images, with the parameter sharing mechanism significantly reducing computational load.76% similarUnverifiedCNN利用参数共享和局部连接大幅减少了参数量，相比传统全连接网络使训练大规模图像识别模型成为可能73% similarUnverifiedCNN通过卷积核在图像上滑动扫描抓取局部特征，再通过池化操作压缩保留最强特征73% similarUnverifiedThe maturity of computer vision is largely attributable to breakthroughs in convolutional neural networks (CNNs).69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/532897API
curl https://kongchang.com/api/v1/knowledge/claims/532897MCP
get_claim(id=532897)