近年來，由於各種安全系統的需求，使得行人偵測(pedestrian detection)技術發展得相當成熟。但是一般的做法都只考慮在單張靜態影像的效能，而忽略了視訊中前一個畫面(frame)可能帶給我們的訊息，使得實際應用上變得沒有效率。本篇論文的目的是要利用關鍵畫面(key-frame)的概念，把物件追蹤(object tracking)的方法結合到影片的行人偵測上。我們將在關鍵畫面偵測得到的行人位置、大小記錄下來，在關鍵畫面間距(interval)時利用這些資訊去做物件追蹤，使得可以快速找到行人的位置、大小等訊息，避免重複搜索時間軸上相鄰的畫面而造成不必要的運算。本文討論的架構大概分成三種，第一種是在單張的關鍵畫面上先偵測行人，然後在後幾張用這些結果作追蹤。第二種是雙向追蹤法，在關鍵畫面做行人偵測後，利用這些結果往前與往後作追蹤，再結合兩個方向的追蹤結果。第三種是可適性的切換模式，先藉由偵測的結果做觀察，等到符合我們設定的條件的時間點，再切換過去追蹤模式。物件追蹤的方法則採用基於梯度與色彩特徵的粒子濾波器(particle filter)，為了可以簡單且快速的找到相同的物體。在我們的實驗結果顯示，雖然這樣的作法造成偵測率(detection rate)些微下降，但是同時會有一定的效率提升。而且現在針對不同物體都適用的偵測方法的研究也越來越多，未來應用到多重物體的系統上，可能會使得效率的改善上更加明顯。

[1] N. Dalal and B. Triggs, ” Histograms of Oriented Gradients for Human Detection,” in IEEE Conference on Computer Vision and Pattern Recognition , San Diego, USA, pages 886 - 893, June 2005.
[2] P. Felzenszwalb, R. Girshick, D. McAllester, “Cascade Object Detection with Deformable Part Models,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2010.
[3] P. Felzenszwalb, R. Girshick, D. McAllester, D. Ramanan, “Object Detection with Discriminatively Trained Part Based Models,” in IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 32, No. 9, September 2010.
[4] P. Felzenszwalb, D. McAllester, D. Ramanan, “A Discriminatively Trained, Multiscale, Deformable Part Model,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2008.
[5] Q. Zhu, S. Avidan, M. Ye and K.-T. Cheng, “Fast Human Detection Using a Cascade of Histograms of Oriented Gradients,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2006.
[6] Y.-T. Chen, C.-S. Chen, Y.-P. Hung, and K.-Y. Chang, “Multi-Class Multi-Instance Boosting for Part-Based Human Detection,” in 9th IEEE International Workshop on Visual Surveillance, VS 2009, October 2009.
[7] Y.-T. Chen and C.-S. Chen, “Fast Human Detection Using A Novel Boosted Cascading Structure with Meta Stages,” in IEEE Transactions on Image Processing, volume 17, number 8, pages 1452-1464, August 2008.
[8] H. Ling and K. Okada, “Diffusion Distance for Histogram Comparison ,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2006.
[9] P. Pérez, C. Hue, J. Vermaak and M. Gangnet, “Color-Based Probabilistic Tracking,” in ECCV, pp. 661–675, Vol. 1, 2002.
[10] B.Babenko, M.-H. Yang and S. Belongie, “Visual Tracking with Online Multiple Instance Learning,” IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2009.
[11] J. Kwon and K. -M. Lee, “Visual Tracking Decomposition,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2010.
[12] S. Walk, N. Majer, K. Schindler and B. Schiele, “New Features and Insights for Pedestrian Detection,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 06/2010, San Francisco, (2010).
[13] C. Wojek, S. Walk and B. Schiele, ”Multi-Cue Onboard Pedestrian Detection,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2009.
[14] M. Andriluka, S. Roth and B. Schiele, “People-Tracking-by-Detection and People-Detection-by-Tracking,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR'08), (2008).
[15] Xiaoyu Wang, Tony X. Han and Shuicheng Yan, “An HOG-LBP Human Detector with Partial Occlusion Handling,” IEEE International Conference on Computer Vision (ICCV 2009), Kyoto, 2009.
[16] M. Enzweiler, A. Eigenstetter, B. Schiele and D. M. Gavrila, “Multi-Cue Pedestrian Classification with Partial Occlusion Handling,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), San Francisco, USA, 2010.
[17] L. Zhu, Y. Chen, A. Yuille and W. Freeman, ”Latent Hierarchical Structural Learning for Object Detection,” in IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2010.
[18] P. Ott and M. Everingham, “Implicit color segmentation features for pedestrian and object detection,” In Proceedings of the IEEE International Conference on Computer Vision (ICCV2009), October 2009.