Beyond Local Search: Tracking Objects Everywhere with Instance-Specific Proposals
會(huì)議：CVPR 2016
實(shí)驗(yàn)室：Australian National University, Hongdong Li
目標(biāo)：以前的方法只能在小范圍內(nèi)查找，本文的方法提供甚至在整張圖上的查找的能力雷厂，實(shí)現(xiàn)跟蹤朽缎。
特色：應(yīng)用edge-based features挖诸，來(lái)自Piotr Doll′ar的一系列工作
A Super-Fast Online Face Tracking System for Video Surveillance
實(shí)驗(yàn)室：自動(dòng)化所
目標(biāo)：快速檢測(cè)監(jiān)控下的多個(gè)人臉斧吐，對(duì)暫時(shí)出畫面的物體魯邦召锈。
方法：KLT + 直方圖驗(yàn)證（保證不是背景）+ 記憶跟蹤
- 直方圖是做在整張臉上的渐扮。
- 開辟一個(gè)buffer用來(lái)存儲(chǔ)跟蹤消失的人臉的模型柑贞。
A Contour-Based Moving Object Detection and Tracking
2005
目標(biāo)：魯棒、快速壳嚎、非剛體物體檢測(cè)和跟蹤
方法：edge-based features(對(duì)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(shí)踐
KCF [1]
- High-Speed Tracking with Kernelized Correlation Filters
- 采用判別式的tracking桐智，需要區(qū)分目標(biāo)和surrounding 環(huán)境，需要大量的訓(xùn)練樣本烟馅，這些樣本之間存在著大量的冗余，于是作者采用創(chuàng)新的circulant matrix來(lái)生成訓(xùn)練樣本然磷，這樣的好處就是得到的數(shù)據(jù)矩陣是circulant郑趁，于是可以利用DFT(離散傅里葉變化)對(duì)角化，從而減少計(jì)算量
- 傅里葉變換可以把循環(huán)矩陣對(duì)角化
- 循環(huán)矩陣是一種特殊形式的 Toeplitz矩陣姿搜，它的行向量的每個(gè)元素都是前一個(gè)行向量各元素依次右移一個(gè)位置得到的結(jié)果寡润。由于可以用離散傅立葉變換快速解循環(huán)矩陣，所以在數(shù)值分析中有重要的應(yīng)用舅柜。
MOSSE[2]
- Matlab上梭纹，對(duì)640*480的圖片不能實(shí)時(shí)
- 但是文章稱在Python using the PyVision library,OpenCV, and SciPy上可以達(dá)到669的幀率
- 通過(guò)仿射變換得到一系列的訓(xùn)練數(shù)據(jù)f和g，計(jì)算所需要的模板h致份。在下一幀变抽，同一個(gè)框內(nèi)，計(jì)算得到最高的響應(yīng)位置就是新的框中心氮块。
“Learning to Track at 100 FPS with Deep Regression Networks”
- http://davheld.github.io/GOTURN/GOTURN.html
- 速度最快的神經(jīng)網(wǎng)絡(luò)跟蹤算法
- ECCV 2016
- 但是在CPU上的速度僅有2.7fps绍载，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]滔蝉，
- Danelljan等人對(duì)基于CF的方法作了改進(jìn)击儡，增加了對(duì)縮放的估計(jì)。
- 聲稱快速并且高效
- 縮放方法用在MOSSE跟蹤方法上蝠引，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(lái)二維圖像的通過(guò)金字塔弄成三維的阳谍，h和g也相應(yīng)變成三維的蛀柴。響應(yīng)最大的那個(gè)層就是scale的最佳值，0.96FPS
LCT[5]
Visual Tracking: An Experimental Survey [6]
- 主要貢獻(xiàn)：systematic analysis and the experimental evaluation of online trackers
- 在130段視頻上進(jìn)行評(píng)測(cè)
- 不評(píng)價(jià)off-line的算法
- 不評(píng)價(jià)contour-based算法矫夯，因?yàn)槌跏蓟容^困難
- 表1,總結(jié)了各種評(píng)價(jià)標(biāo)準(zhǔn)
- F score:
一些數(shù)據(jù)庫(kù)
- 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(píng)價(jià)標(biāo)準(zhǔn)
- PETS：Performance Evaluation of Tracking and Surveillance
- PETS and VACE名扛，CLEAR：for evaluating the performance of multiple target detection and tracking
參考文獻(xiàn)
[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.
