深度网络推理加速（Towards Lightweight Convolutional Neural Networks for Object Detection）
摘要：本文研究目标是少类别实时目标检测，研究了在保留较高检测率的前提下最大程度的减少模型大小，最终实现了在CPU上的实时检测。本文的推理加速机制和量化压缩等方法是能够并行的。算法 1， 经过增大feature map的尺寸，减少通道数能够获得准确率且快的检测模型，这么作的依据是对于实际的不多类别的目标检测问题，用于解决多类别分类问题的网络中不少通道是冗余的。网络 2， 经过在大的预训练模型上简
相关文章
- 1. 论文笔记: PVANet: Lightweight Deep Neural Networks for Real-time Object Detection
- 2. Convolutional Neural Networks（week 3）---Object detection
- 3. 【PVANet】《PVANET：Deep but Lightweight Neural Networks for Real-time Object Detection》
- 4. 【目标检测】PVANET:Deep but Lightweight Neural Networks for Real-time Object Detection
- 5. Deep Neural Networks for Object Detection
- 6. PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
- 7. Convolutional Neural Network - Object Detection
- 8. Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages
- 9. ROLO:Spatially Supervised Recurrent Convolutional Neural Networks for Visual Object Tracking
- 10. LCDet Low-Complexity Fully-Convolutional Neural Networks for Object Detection in Embedded Systems
- 更多相关文章...
- • Docker 镜像加速 - Docker教程
- • 网络协议是什么？ - TCP/IP教程
- • 算法总结-深度优先算法
- • 使用阿里云OSS+CDN部署前端页面与加速静态资源
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。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. 论文笔记: PVANet: Lightweight Deep Neural Networks for Real-time Object Detection
- 2. Convolutional Neural Networks（week 3）---Object detection
- 3. 【PVANet】《PVANET：Deep but Lightweight Neural Networks for Real-time Object Detection》
- 4. 【目标检测】PVANET:Deep but Lightweight Neural Networks for Real-time Object Detection
- 5. Deep Neural Networks for Object Detection
- 6. PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
- 7. Convolutional Neural Network - Object Detection
- 8. Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages
- 9. ROLO:Spatially Supervised Recurrent Convolutional Neural Networks for Visual Object Tracking
- 10. LCDet Low-Complexity Fully-Convolutional Neural Networks for Object Detection in Embedded Systems