Learning to Count with CNN Boosting

In this paper, we address the task of object counting in images. We follow modern learning approaches in which a density map is estimated directly from the input image. We employ CNNs and incorporate two significant improvements to the state of the art methods: layered boosting and selective sampling. As a result, we manage both to increase the counting accuracy and to reduce processing time. Moreover, we show that the proposed method is effective, even in the presence of labeling errors. Extensive experiments on five different datasets demonstrate the efficacy and robustness of our approach. Mean Absolute error was reduced by 20 % to 35 %. At the same time, the training time of each CNN has been reduced by 50 %.

[1]  Hironobu Fujiyoshi,et al.  Improving Quality of Training Samples Through Exhaustless Generation and Effective Selection for Deep Convolutional Neural Networks , 2015, VISAPP.

[2]  Ullrich Köthe,et al.  Learning to count with regression forest and structured labels , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[3]  Ming Yang,et al.  Web-scale training for face identification , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[4]  Nassir Navab,et al.  Robust Optimization for Deep Regression , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[5]  Ce Liu,et al.  Depth Extraction from Video Using Non-parametric Sampling , 2012, ECCV.

[6]  Lei Wang,et al.  A study of AdaBoost with SVM based weak learners , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[7]  Shaogang Gong,et al.  Feature Mining for Localised Crowd Counting , 2012, BMVC.

[8]  Xuming He,et al.  Discrete-Continuous Depth Estimation from a Single Image , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Xiaochun Cao,et al.  Deep People Counting in Extremely Dense Crowds , 2015, ACM Multimedia.

[10]  Shaogang Gong,et al.  Cumulative Attribute Space for Age and Crowd Density Estimation , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[11]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[12]  Andrew Zisserman,et al.  Learning To Count Objects in Images , 2010, NIPS.

[13]  Xiaogang Wang,et al.  Cross-scene crowd counting via deep convolutional neural networks , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Rocco A. Servedio,et al.  Random classification noise defeats all convex potential boosters , 2008, ICML.

[15]  J. Friedman.  Stochastic gradient boosting , 2002 .

[16]  Svetha Venkatesh,et al.  Efficient algorithms for subwindow search in object detection and localization , 2009, CVPR 2009.

[17]  Nuno Vasconcelos,et al.  Privacy preserving crowd monitoring: Counting people without people models or tracking , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[18]  Ian D. Reid,et al.  Learning Depth from Single Monocular Images Using Deep Convolutional Neural Fields , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Visvanathan Ramesh,et al.  Fast Crowd Segmentation Using Shape Indexing , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[20]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[21]  Meng Wang,et al.  Deep Learning of Scene-Specific Classifier for Pedestrian Detection , 2014, ECCV.

[22]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[23]  Xiaogang Wang,et al.  Fully Convolutional Neural Networks for Crowd Segmentation , 2014, ArXiv.

[24]  Sridha Sridharan,et al.  Crowd Counting Using Multiple Local Features , 2009, 2009 Digital Image Computing: Techniques and Applications.

[25]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[26]  Yoshua Bengio,et al.  Deep Sparse Rectifier Neural Networks , 2011, AISTATS.

[27]  Ryuzo Okada,et al.  COUNT Forest: CO-Voting Uncertain Number of Targets Using Random Forest for Crowd Density Estimation , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[28]  Gang Hua,et al.  A convolutional neural network cascade for face detection , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[29]  Tomaso A. Poggio,et al.  Example-Based Learning for View-Based Human Face Detection , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[30]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1995, EuroCOLT.

[31]  Stefano Soatto,et al.  Boosting Convolutional Features for Robust Object Proposals , 2015, ArXiv.

[32]  Haroon Idrees,et al.  Multi-source Multi-scale Counting in Extremely Dense Crowd Images , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[33]  Tsuhan Chen,et al.  Toward Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[34]  Xiaogang Wang,et al.  Multi-stage Contextual Deep Learning for Pedestrian Detection , 2013, 2013 IEEE International Conference on Computer Vision.

[35]  Ashutosh Saxena,et al.  Learning Depth from Single Monocular Images , 2005, NIPS.