Learning to Track at 100 FPS with Deep Regression Networks ECCV 2016 论文笔记
2017-03-22 15:15
603 查看
Learning to Track at 100 FPS with Deep Regression Networks ECCV 2016 论文笔记
工程网页:http://davheld.github.io/GOTURN/GOTURN.htmlGitHub 地址:https://github.com/davheld/GOTURN
本文第一次利用 Deep Learning 技术将跟踪的速度维持在 100fps,当然是使用 GPU 的前提下。本文的流程框架如下所示:
将跟踪看做是回归问题,直接根据上一帧的位置,回归出当前帧的location。类比于 基于Siamese 网络的匹配,仅用第一帧作为 target object,本文方法不需要提候选的 proposal,直接进行 bounding box 的回归。很好的避开了 CNN 在跟踪问题上速度慢的难题。
是的,没了,就这么多。这就是文章的主要思想了。。。
另外:给些参考的blog,因为他们讲的更加详细。
1. http://blog.csdn.net/cuclxt/article/details/51570255
2. http://blog.csdn.net/autocyz/article/details/52648776
3. https://zhuanlan.zhihu.com/p/22715531 (强烈推荐)
相关文章推荐
- Learning to Track at 100 FPS with Deep Regression Networks 论文理解及应用笔记(二)
- Learning to Track at 100 FPS with Deep Regression Networks 论文笔记
- Learning to Track at 100 FPS with Deep Regression Networks 论文理解及应用笔记(一)
- 计算机视觉之目标跟踪——论文Learning to Track at 100 FPS with Deep Regression Networks
- Learning to Track at 100 FPS with Deep Regression Networks
- 深度学习Tracking(1)——Learning to Track at 100 FPS with Deep Regression Networks(代码理解)
- 深度学习Tracking(1)——Learning to Track at 100 FPS with Deep Regression Networks
- 论文笔记:MDPTracking,Learning to Track: Online Multi-Object Tracking by Decision Making
- GOTURN——Learning to Track at 100 FPS with Deep Regression Networks
- 目标跟踪之GOTURN:Learning to Track at 100 FPS with Deep Regression Networks
- 【目标跟踪】Learning to Track at 100 FPS with Deep Regression Networks 学习
- GOTURN: Learning to Track at 100 FPS with Deep Regression Networks笔记
- 论文笔记之:Learning to Track: Online Multi-Object Tracking by Decision Making
- 【论文笔记】From Facial Parts Responses to Face Detection: A Deep Learning Approach
- [论文笔记] Learning to Read Chest X-Rays Recurrent Neural Cascade Model for Automated Image Annotation
- Sequence to Sequence Learning with Neural Networks论文笔记
- ICLR 2016 - Workshop Track International Conference on Learning Representations 论文papers
- 论文笔记 Ensemble of Deep Convolutional Neural Networks for Learning to Detect Retinal Vessels in Fundus
- ICLR 2016 - Workshop Track International Conference on Learning Representations 论文papers
- 机器学习_论文笔记_1: A few useful things to know about machine learning