【DeepLearning】简述Visual Tracking with Fully Convolutional Networks
Visual Tracking with Fully Convolutional Networks 背景 研究意义 整体结构 算法实现 实验结果 一、背景： 1.现有的跟踪方法多数采用生成或区分模型，并将不同类别的但同时发生的对象分开。 2.它们依赖于低级别的手工特征，这些特征不能捕捉到目标的语义信息，对外观显著变化的物体效果不好，只有有限的区分能力。 3.在处理重大外观变化，姿态变化，严重遮挡和
相关文章
- 1. Fully-Convolutional Siamese Networks for Object Tracking
- 2. visual tracking 经典文献翻译（1）- Fully-Convolutional Siamese Networks for Object Tracking
- 3. Fast Image Processing with Fully-Convolutional Networks
- 4. ROLO:Spatially Supervised Recurrent Convolutional Neural Networks for Visual Object Tracking
- 5. Fully Convolutional Networks
- 6. MDNet: Learning Multi-Domain Convolutional Neural Networks for Visual Tracking
- 7. Automatic skin lesion segmentation with fully convolutional-deconvolutional networks
- 8. Multi-Oriented Text Detection with Fully Convolutional Networks
- 9. Deeper Depth Prediction with Fully Convolutional Residual Networks
- 10. 【Tracking 系列：一】Siamese-FC｜Fully-Convolutional Siamese Networks for Object Tracking
- 更多相关文章...
- • Scala 简介 - Scala教程
- • AJAX 简介 - PHP教程
- • Github 简明教程
- • Java Agent入门实战（二）-Instrumentation源码概述
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. Fully-Convolutional Siamese Networks for Object Tracking
- 2. visual tracking 经典文献翻译（1）- Fully-Convolutional Siamese Networks for Object Tracking
- 3. Fast Image Processing with Fully-Convolutional Networks
- 4. ROLO:Spatially Supervised Recurrent Convolutional Neural Networks for Visual Object Tracking
- 5. Fully Convolutional Networks
- 6. MDNet: Learning Multi-Domain Convolutional Neural Networks for Visual Tracking
- 7. Automatic skin lesion segmentation with fully convolutional-deconvolutional networks
- 8. Multi-Oriented Text Detection with Fully Convolutional Networks
- 9. Deeper Depth Prediction with Fully Convolutional Residual Networks
- 10. 【Tracking 系列：一】Siamese-FC｜Fully-Convolutional Siamese Networks for Object Tracking