Abstract
Generally the performance of a pedestrian detector will decrease rapidly, when it is trained on a fixed training set but applied to specific scenes. The reason is that in the training set only a few samples are useful for the specific scenes while other samples may disturb the accurate detections. Traditional methods solve this problem by transfer learning which suffer the problem of keeping source samples or artificially labeling a few samples in the detection phase. In this paper, we propose a new method to bypass these defects by predicting pedestrian classifier for each sample in the detection phase. A classifier regression model is trained in the source domain in which each sample has a proprietary classifier. In the detection phase, a pedestrian classifier is predicted for each candidate window in an image. Thus, for the samples in the target domain, the pedestrian classifiers are different. Our main contributions are: (1) a new adaptive detector without keeping source samples or labeling a few new target samples; (2) a new dimensionality reduction method for classifier vector which simultaneously ensures the performance of both reconstruction and classification; (3) a two-stage regression neural model which can handle the high-dimensional regression problem effectively. Experiments prove that our method can achieve the state-of-the-art results on two pedestrian datasets.
Similar content being viewed by others
Notes
Download link: http://vision.ucsd.edu/~pdollar/toolbox/.
References
Andriluka M, Roth S, Schiele B (2009) Pictorial structures revisited: people detection and articulated pose estimation. In: Conference on computer vision and pattern recognition (CVPR), pp 1014–1021
Caseiro R, Henriques JF, Martins P, Batista J (2015) Beyond the shortest path: unsupervised domain adaptation by sampling subspaces along the spline flow. In: Conference on computer vision and pattern recognition (CVPR), pp 3846–3854
Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. In: Conference on computer vision and pattern recognition (CVPR), pp 886–893
Dollár P, Appel R, Belongie S, Perona P (2014) Fast feature pyramids for object detection. TIEEE Tran Pattern Anal Mach Intell (PAMI) 36(8):1532–1545
Everingham M, Van Gool L, Williams CK, Winn J, Zisserman A (2010) The pascal visual object classes (voc) challenge. Int J Comput Vis (IJCV) 88(2):303–338
Felzenszwalb PF, Girshick RB, McAllester D, Ramanan D (2010) Object detection with discriminatively trained part-based models. TIEEE Trans Pattern Anal Mach Intell (PAMI) 32(9):1627–1645
Gall J, Lempitsky V (2013) Class-specific hough forests for object detection. In: Conference on computer vision and pattern recognition (CVPR), pp 143–157
Girshick R, Donahue J, Darrell T, Malik, J (2014a) Rich feature hierarchies for accurate object detection and semantic segmentation. In: Conference on computer vision and pattern recognition (CVPR), pp 580–587
Girshick R, Donahue J, Darrell T, Malik J (2014b) Rich feature hierarchies for accurate object detection and semantic segmentation. In: Conference on computer vision and pattern recognition (CVPR), pp 580–587
Gong B, Shi Y, Sha F, Grauman K (2012) Geodesic flow kernel for unsupervised domain adaptation. In: Conference on computer vision and pattern recognition (CVPR), pp 2066–2073
Gopalan R, Li R, Chellappa R (2011) Domain adaptation for object recognition: an unsupervised approach. In: International conference on computer vision (ICCV), pp 999–1006
Gould S, Fulton R, Koller, D (2009) Decomposing a scene into geometric and semantically consistent regions. In: International Conference on computer vision (ICCV), pp 1–8
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In: International conference on computer vision (ICCV)
Jiang W, Zavesky E, Chang SF, Loui A (2008) Cross-domain learning methods for high-level visual concept classification. In: International conference on image processing (ICIP), pp 161–164
Kate S, Brian K, Mario F, Trevor D (2010) Adapting visual category models to new domains. In: European conference on computer vision (ECCV), pp 213–226
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems (NIPS), pp 1097–1105
Li X, Ye M, Fu M, Xu P, Li T (2015) Domain adaption of vehicle detector based on convolutional neural networks. IJCAS 13(4):1020–1031
Malisiewicz T, Gupta A, Efros A et al. (2011) Ensemble of exemplar-svms for object detection and beyond. In: International conference on computer vision (ICCV), pp 89–96
Nair V, Clark JJ (2004) An unsupervised, online learning framework for moving object detection. In: Conference on computer vision and pattern recognition (CVPR), pp 317–324
Oren M, Papageorgiou C, Sinha P, Osuna E, Poggio T (1997) Pedestrian detection using wavelet templates. In: Conference on computer vision and pattern recognition (CVPR), pp 193–99
Overett G, Petersson L, Brewer N, Andersso, L, Pettersson N (2008) A new pedestrian dataset for supervised learning. In: IVS, pp 373–378
Pang J, Huang Q, Yan S, Jiang S, Qin L (2011) Transferring boosted detectors towards viewpoint and scene adaptiveness. IEEE Trans Image Process (TIP) 20(5):1388–1400
Schulter S, Leistner C, Wohlhart P, Roth PM, Bischof H (2014) Accurate object detection with joint classification-regression random forests. In: Conference on computer vision and pattern recognition (CVPR), pp 923–930
Sermanet P, Kavukcuoglu K, Chintala S, LeCun Y (2013) Pedestrian detection with unsupervised multi-stage feature learning. In: Conference on computer vision and pattern recognition (CVPR), pp 3626–3633
Sun Y, Wang X, Tang X (2013) Deep convolutional network cascade for facial point detection. In: Conference on computer vision and pattern recognition (CVPR), pp 3476–3483
Tian Y, Luo P, Wang X, Tang X (2015) Pedestrian detection aided by deep learning semantic tasks. In: Conference on computer vision and pattern recognition (CVPR), pp 5079–5087
Wang M, Li W, Wang X (2012) Transferring a generic pedestrian detector towards specific scenes. In: Conference on computer vision and pattern recognition (CVPR), pp 3274–3281
Wang M, Wang X (2011) Automatic adaptation of a generic pedestrian detector to a specific traffic scene. In: Conference on computer vision and pattern recognition (CVPR), pp 3401–3408
Wang X, Wang M, Li W (2014) Scene-specific pedestrian detection for static video surveillance. TIEEE Trans Pattern Anal Mach Intell (PAMI) 36(2):361–374
Wu Y, Wang L, Cui F, Zhai H, Dong B, Wang JY (2016) Cross-model convolutional neural network for multiple modality data representation. Neural Comput Appl. doi:10.1007/s00521-016-2824-4
Yang J, Yan R, Hauptmann AG (2007) Cross-domain video concept detection using adaptive svms. In: ACM international conference on multimedia (ACMM), pp 188–197
Yin Z, Kong D, Shao G, Ning X, Jin W, Wang JY (2016) A-optimal convolutional neural network. Neural Comput Appl. doi:10.1007/s00521-016-2783-9
Zeng X, Ouyang W, Wang M, Wang X (2014) Deep learning of scene-specific classifier for pedestrian detection. In: European conference on computer vision (ECCV), pp 472–487
Zeng X, Ouyang W, Wang X (2013) Multi-stage contextual deep learning for pedestrian detection. In: International conference on computer vision (ICCV), pp 121–128
Zhang H, Cao X, Ho JK, Chow TW (2017) Object-level video advertising: an optimization framework. IEEE Trans Ind Inform 13(2):520–531
Zhang H, Li J, Ji Y, Yue H (2017) Understanding subtitles by character-level sequence-to-sequence learning. IEEE Trans Ind Inform 13(2):616–624
Acknowledgements
This work was supported in part by the National Natural Science Foundation of China (61375038) and Applied Basic Research Programs of Sichuan Science and Technology Department (2016JY0088).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled, Adaptive Pedestrian Detection by Predicting Classifier.
Rights and permissions
About this article
Cite this article
Tang, S., Ye, M., Xu, P. et al. Adaptive pedestrian detection by predicting classifier. Neural Comput & Applic 31, 1189–1200 (2019). https://doi.org/10.1007/s00521-017-3152-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-017-3152-z