Beyond Local Search: Tracking Objects Everywhere with Instance-Specific Proposals
會議：CVPR 2016
實驗室：Australian National University, Hongdong Li
目標：以前的方法只能在小范圍內查找，本文的方法提供甚至在整張圖上的查找的能力，實現跟蹤。
特色：應用edge-based features，來自Piotr Doll′ar的一系列工作
A Super-Fast Online Face Tracking System for Video Surveillance
實驗室：自動化所
目標：快速檢測監控下的多個人臉，對暫時出畫面的物體魯邦。
方法：KLT + 直方圖驗證（保證不是背景）+ 記憶跟蹤
- 直方圖是做在整張臉上的。
- 開辟一個buffer用來存儲跟蹤消失的人臉的模型。
A Contour-Based Moving Object Detection and Tracking
2005
目標：魯棒、快速、非剛體物體檢測和跟蹤
方法：edge-based features(對光照不敏感) + 梯度光流法（gradient-based optical flow technique）
Face Tracking: An implementation of the Kanade-Lucas-Tomasi Tracking algorithm
KLT在人臉跟蹤上的實踐
KCF [1]
- High-Speed Tracking with Kernelized Correlation Filters
- 采用判別式的tracking，需要區分目標和surrounding 環境，需要大量的訓練樣本，這些樣本之間存在著大量的冗余，于是作者采用創新的circulant matrix來生成訓練樣本，這樣的好處就是得到的數據矩陣是circulant，于是可以利用DFT(離散傅里葉變化)對角化，從而減少計算量
- 傅里葉變換可以把循環矩陣對角化
- 循環矩陣是一種特殊形式的 Toeplitz矩陣，它的行向量的每個元素都是前一個行向量各元素依次右移一個位置得到的結果。由于可以用離散傅立葉變換快速解循環矩陣，所以在數值分析中有重要的應用。
MOSSE[2]
- Matlab上，對640*480的圖片不能實時
- 但是文章稱在Python using the PyVision library,OpenCV, and SciPy上可以達到669的幀率
- 通過仿射變換得到一系列的訓練數據f和g，計算所需要的模板h。在下一幀，同一個框內，計算得到最高的響應位置就是新的框中心。
“Learning to Track at 100 FPS with Deep Regression Networks”
- http://davheld.github.io/GOTURN/GOTURN.html
- 速度最快的神經網絡跟蹤算法
- ECCV 2016
- 但是在CPU上的速度僅有2.7fps，不能容忍
MDnet[3]，
- Learning Multi-Domain Convolutional Neural Networks for Visual Tracking
- VOT 2015冠軍
- http://cvlab.postech.ac.kr/research/mdnet/
DSST[4]，
- Danelljan等人對基于CF的方法作了改進，增加了對縮放的估計。
- 聲稱快速并且高效
- 縮放方法用在MOSSE跟蹤方法上，但是該縮放方法可以普遍用于其他跟蹤方法
- 其fast scale search速度為：24 fps
- 提出的Exhaustive Scale Space Tracking就是將原來二維圖像的通過金字塔弄成三維的，h和g也相應變成三維的。響應最大的那個層就是scale的最佳值，0.96FPS
LCT[5]
Visual Tracking: An Experimental Survey [6]
- 主要貢獻：systematic analysis and the experimental evaluation of online trackers
- 在130段視頻上進行評測
- 不評價off-line的算法
- 不評價contour-based算法，因為初始化比較困難
- 表1,總結了各種評價標準
- F score:
一些數據庫
- OTB50 http://cvlab.hanyang.ac.kr/tracker_benchmark/datasets.html
- OTB100
- VOT2014[7] http://www.votchallenge.net/
- VOT2015
- BoBoT dataset：“D. A. Klein, D. Schulz, S. Frintrop, and A. B. Cremers, “Adaptive
- real-time video-tracking for arbitrary objects,” in Proc. IEEE IROS, Taipei, Taiwan, 2010, pp. 772–777.”
- CAVIAR dataset：few but long and difficult videos
- i-LIDS Multiple-Camera Tracking Scenario
- 3DPeS dataset：contains videos with more than 200 people walking as recorded from eight different cameras in very long video sequences
- PETS-series：
- TRECVid video dataset：large video benchmark
- ALOV++ dataset：proposed by [6]; more than 300 video sequences; http://crcv.ucf.edu/data/ALOV++/
評價標準
- PETS：Performance Evaluation of Tracking and Surveillance
- PETS and VACE，CLEAR：for evaluating the performance of multiple target detection and tracking
參考文獻
[1] J. F. Henriques, R. Caseiro, P. Martins, and J. Batista, "High-Speed Tracking with Kernelized Correlation Filters," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 37, pp. 583-596, 2015.
[2] D. S. Bolme, J. R. Beveridge, B. A. Draper, and Y. M. Lui, "Visual object tracking using adaptive correlation filters," in 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010, pp. 2544-2550.
[3] H. Nam and B. Han. (2015, October 1, 2015). Learning Multi-Domain Convolutional Neural Networks for Visual Tracking. ArXiv e-prints 1510. Available: http://adsabs.harvard.edu/abs/2015arXiv151007945N
[4] M. Danelljan, G. H?ger, F. Khan, and M. Felsberg, "Accurate scale estimation for robust visual tracking," in British Machine Vision Conference, Nottingham, September 1-5, 2014, 2014.
[5] C. Ma, X. Yang, Z. Chongyang, and M. H. Yang, "Long-term correlation tracking," in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 5388-5396.
[6] A. W. M. Smeulders, D. M. Chu, R. Cucchiara, S. Calderara, A. Dehghan, and M. Shah, "Visual Tracking: An Experimental Survey," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, pp. 1442-1468, 2014.
[7] M. Kristan, J. Matas, A. Leonardis, T. Vojí?, R. Pflugfelder, G. Fernández, G. Nebehay, F. Porikli, and L. ?ehovin, "A Novel Performance Evaluation Methodology for Single-Target Trackers," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 38, pp. 2137-2155, 2016.
