[论文笔记]TextBoxes A Fast Text Detector with a Single Deep Neural Network
TextBoxes A Fast Text Detector with a Single Deep Neural Network 相比于ssd，创新点如下： 1.改变生成的prior boxes的aspect ratios，改为1,2,3,5,7和10。并且设置了vertical offsets 2.将３×３的卷积核改为1×５，改善了特征提取效果。 3.测试阶段：将输入的image重新变形为5种大
相关文章
- 1. 论文笔记：TextBoxes: A Fast Text Detector with a Single Deep Neural Network
- 2. 《TextBoxes: A Fast Text Detector with a Single Deep Neural Network》阅读翻译
- 3. 【Paper Reading】【TextBoxess: A Fast Text Detector with a Single Deep Neural Network】
- 4. 《TextBoxes++: A Single-Shot Oriented Scene Text Detector》论文笔记
- 5. 论文阅读 | Towards a Robust Deep Neural Network in Text Domain A Survey
- 6. Explaining How a Deep Neural Network Trained with End-to-End Learning Steers a Car论文笔记
- 7. 【论文阅读笔记】Deep Neural Network Compression with Single and Multiple Level Quantization
- 8. 【论文解读】Pixel-Anchor: A Fast Oriented Scene Text Detector with Combined Networks
- 9. 论文阅读：Fast Single Image Rain Removal via a Deep Decomposition-Composition Network
- 10. 翻译和笔记--FOTS: Fast Oriented Text Spotting with a Unified Network
- 更多相关文章...
- • SVG
- SVG 教程 - • CAP理论是什么？ - NoSQL教程
- • Tomcat学习笔记（史上最全tomcat学习笔记）
- • Scala 中文乱码解决
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
最新文章
欢迎关注本站公众号,获取更多信息
相关文章
- 1. 论文笔记：TextBoxes: A Fast Text Detector with a Single Deep Neural Network
- 2. 《TextBoxes: A Fast Text Detector with a Single Deep Neural Network》阅读翻译
- 3. 【Paper Reading】【TextBoxess: A Fast Text Detector with a Single Deep Neural Network】
- 4. 《TextBoxes++: A Single-Shot Oriented Scene Text Detector》论文笔记
- 5. 论文阅读 | Towards a Robust Deep Neural Network in Text Domain A Survey
- 6. Explaining How a Deep Neural Network Trained with End-to-End Learning Steers a Car论文笔记
- 7. 【论文阅读笔记】Deep Neural Network Compression with Single and Multiple Level Quantization
- 8. 【论文解读】Pixel-Anchor: A Fast Oriented Scene Text Detector with Combined Networks
- 9. 论文阅读：Fast Single Image Rain Removal via a Deep Decomposition-Composition Network
- 10. 翻译和笔记--FOTS: Fast Oriented Text Spotting with a Unified Network