Very Deep Convolutional Networks for Large-Scale Image Recognition

• 类型：arxiv
• 标识：1409.1556
• 链接：https://arxiv.org/abs/1409.1556
• 主题：evaluation
• 主分类：multimodal
• 形态：method
• 被引：113645
• 被引来源：Semantic Scholar
• S2被引：113645
• OpenAlex被引：75491
• 影响力被引：14619
• TLDR：This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
• OpenAlex ID：W1686810756
• OpenAlex DOI：10.48550/arxiv.1409.1556
• DOI：10.48550/arxiv.1409.1556
• DOI来源：OpenAlex
• 开放获取：green
• 开放获取链接：https://arxiv.org/pdf/1409.1556
• OpenAlex更新：2026-08-25
• 待LLM分类：否
• 成熟度：production
• 场景：image classification、CNN、feature extraction
• 标题中文：Very Deep Convolutional Networks for Large-Scale Image Recognition
• TLDR中文：本文研究了在采用极小卷积滤波器的架构下，卷积网络深度对大规模图像识别精度的影响，并表明将深度推进至 16-19 个权重层，可在先前 SOTA 配置基础上取得显著提升。
• 来源文件：
• [OpenAlex discover]
• [S2 enrich]