1.深度学习的目标跟踪算法综述_卢湖川
2.Large Margin Object Tracking with Circulant Feature Maps(APCE)
1.高峰检测
2.多置信度更新
(LMCF置信度指标是平均峰值相关能量(average peak-to correlation energy, APCE))
(置信度指标还有MOSSE中的峰值旁瓣比(Peak to Sidelobe Ratio, PSR))
(CSR-DCF的空域可靠性)
3.fully-convolutional siamese networks for object tracking (siamesefc)
5个尺度(58 FPS) 3个尺度(86FPS)
1.数据及大(4500个视频)
2.卷积特征
4.搜索区域大(4倍)
//srdcf(速度慢)
LMCF
BACF
staple
eco
siamesefc
yolo
fastercnn
mdnet
goturn
maskrcnn
总结
//srdcf(速度慢)
ICCV,ECCV,CVPR
HuChuan Lu
特征选择(HOG,CNN,CN)
分类器(SVM,岭回归,Deep Netural NetWorks)
模型更新(Fixed interval,High-condifience strategy)
入门
1.Uderstading and diagnoising visual tracking systems
2.KCF
3.Learning multi-domain convotional netural networks for visual tracking(Mdnet)
网址: https://github.com/foolwood/benchmark_results