Local structured representation for generic object detection

Abstract

Structure information plays an important role in both object recognition and detection. This paper studies what visual structure is and addresses the problem of structure modeling and representation from two aspects: visual feature and topology model. Firstly, at feature level, we propose Local Structured Descriptor to capture the object’s local structure effectively, and develop the descriptors from shape and texture information, respectively. Secondly, at topology level, we present a local structured model with a boosted feature selection and fusion scheme. All experiments are conducted on the challenging PASCAL Visual Object Classes (VOC) datasets from VOC2007 to VOC2010. Experimental results show that our method achieves very competitive performance.

Keywords

Local Structured Descriptor / Local Structured Model / Object Representation / Object Structure / Object Detection / PASCAL VOC

References

[1] AlexeB, Deselaers T, FerrariV . Measuring the objectness of image windows. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(11): 2189–2202

[2] ChengM M, ZhangZ, LinW Y, Torr P. Bing: binarized normed gradients for objectness estimation at 300fps. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2014, 3286–3293

[3] ZitnickC, Dollár P. Edge boxes: locating object proposals from edges. In: Proceedings of European Conference on Computer Vision. 2014, 391–405

[4] YaoC, BaiX, LiuW. A unified framework for multioriented text detection and recognition. IEEE Transactions on Image Processing, 2014, 23(11): 4737–4749

[5] ZhuY, YaoC, BaiX. Scene text detection and recognition: recent advances and future trends. Frontiers of Computer Science, 2016, 10(1): 19–36

[6] DalalN, TriggsB. Histograms of oriented gradients for human detection. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2005, 886–893

[7] VedaldiA, Gulshan V, VarmaM , ZissermanA. Multiple kernels for object detection. In: Proceedings of IEEE International Conference on Computer Vision. 2009, 606–613

[8] WangX, HanT X, YanS. An HOG-LBP human detector with partial occlusion handling. In: Proceedings of IEEE International Conference on Computer Vision. 2009, 32–39

[9] FelzenszwalbP, Girshick R, McAllesterD , RamananD. Object detection with discriminatively trained part-based models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(9): 1627–1645

[10] FergusR, PeronaP, ZissermanA. Object class recognition by unsupervised scale-invariant learning. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2003, 264–271

[11] SchnitzspanP, RothS, SchieleB. Automatic discovery of meaningful object parts with latent CRFs. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2010, 121–128

[12] YangY, Ramanan D. Articulated pose estimation with flexible mixtures-of-parts. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2011, 1385–1392

[13] ZhuL, ChenY, YuilleA L, Freeman W T. Latent hierarchical structural learning for object detection. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2010, 1062–1069

[14] FischlerM, Elschlager R. The representation and matching of pictorial structures. IEEE Transactions on Computers, 1973, 22(1): 67–92

[15] OjalaT, Pietikäinen M, HarwoodD . A comparative study of texture measures with classification based on featured distributions. Pattern Recognition, 1996, 29(1): 51–59

[16] LoweD G.Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 2004, 60(2): 91–110

[17] MarkE, GoolL, WilliamsC K , WinnJ, Zisserman A. The pascal visual object classes (VOC) challenge. International Journal of Computer Vision, 2010, 88(2): 303–338

[18] ZhangJ, HuangK, YuY, TanT. Boosted local structured HOG-LBP for object localization. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2011, 1393–1400

[19] Papageorgiou, C, Poggio T. A trainable system for object detection.International Journal of Computer Vision, 2000, 38(1): 15–33

[20] ViolaP, JonesM J. Robust real-time face detection. International Journal of Computer Vision, 2004, 57(2): 137–154

[21] LeeT S. Image representation using 2D gabor wavelets. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1996, 18(10): 959–971

[22] ShechtmanE, IraniM. Matching local self-similarities across images and videos. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2007, 1–8

[23] FerrariV, Fevrier L, JurieF , SchmidC. Groups of adjacent contour segments for object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(1): 36–51

[24] BaiX, BaiS, ZhuZ, Latecki L J. 3D shape matching via two layer coding. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(12): 2361–2373

[25] LazebnikS, SchmidC, PonceJ. Beyond bags of features: spatial pyramid matching for recognizing natural scene categories. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2006, 2169–2178

[26] SivicJ, Russell B, EfrosA , ZissermanA, Freeman W. Discovering objects and their location in images. In: Proceedings of IEEE International Conference on Computer Vision. 2005, 370–377

[27] FelzenszwalbP F, Huttenlocher D P. Distance transforms of sampled functions. Theory of Computing, 2012, 8(1): 415–428

[28] EsteparR S J. Local Structure tensor for multidimensional signal processing: applications to medical image analysis. Dissertation for the Doctoral Degree. Valladolid:University of Valladolid, 2005

[29] MorroneC, BurrD. Feature detection in human vision: a phasedependent energy model. In: Proceedings of the Royal Society of London B: Biological Sciences. 1988, 221–245

[30] VenkateshS, OwensR. On the classification of image features. Pattern Recognition Letters, 1990, 11(5): 339–349

[31] GranlundG H, Knutsson H. Signal Processing for Computer Vision. Dordrecht: Kluwer Academic Publishers, 1995

[32] OlivaA, Torralba A. Modeling the shape of the scene: a holistic representation of the spatial envelope. International Journal of Computer Vision, 2001, 42(3): 145–175

[33] OjalaT, Pietikainen M, MaenpaaT . Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7): 971–987

[34] VarmaM, BabuB R. More generality in efficient multiple kernel learning. In: Proceedings of International Conference onMachine Learning. 2009, 1065–1072

[35] FriedmanJ, HastieT, TibshiraniR . Additive logistic regression: a statistical view of boosting. Annuals of Statistics, 2000, 28(2): 374–376

[36] HussainS, TriggsB. Feature sets and dimensionality reduction for visual object detection. In: Proceedings of British Machine Vision Conference. 2010

[37] FelzenszwalbP F, Girshick R B, McAllesterD . Discriminatively Trained Deformable Part Models, Release 3

[38] Felzenszwalb, P F, Girshick R B, McAllesterD . Discriminatively Trained Deformable Part Models, Release 4, 2010

[39] GehlerP, Nowozin S. On feature combination for multiclass object classification. In: Proceedings of IEEE International Conference on Computer Vision. 2009, 221–228

[40] TorralbaA, MurphyK, FreemanW. Sharing features: efficient boosting procedures for multiclass object detection. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2004, 762–769

[41] EveringhamM, GoolV L, WilliamsC K I , WinnJ, Zisserman A. Empirical analysis of detection cascades of boosted classifiers for rapid object detection. Lecture Notes in Computer Science, 2003, 2781: 297–304

[42] EveringhamM, GoolV L, WilliamsC K I , WinnJ, Zisserman A. The PASCAL visual object classes challenge 2007 (VOC2007) results. International Journal of Computer Vision, 2010, 88(2): 303–338

[43] DesaiC, Ramanan D, FowlkesC . Discriminative models for multiclass object layout. In: Proceedings of IEEE International Conference on Computer Vision. 2009, 229–236

[44] PedersoliM, Vedaldi A, GonzalezJ . A coarse-to-fine approach for fast deformable object detection. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2011, 1353–1360

[45] RazaviN, GallJ, GoolV L. Scalable multi-class object detection. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2011, 1505–1512

[46] DivvalaS K, Zitnick C, KapoorA , BakerS. Detecting objects using unsupervised parts-based attributes. Technical Report CMU-RI-TR-11- 10, Robotics Institute. 2010

[47] SchnitzspanP, FritzM, RothS, Schiele B. Discriminative structure learning of hierarchical representations for object detection. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2009, 2238–2245

[48] MalisiewiczT, GuptaA, EfrosA A. Ensemble of exemplar-svms for object detection and beyond. In: Proceedings of IEEE International Conference on Computer Vision. 2011, 89–96

[49] DuboutC, Fleuret F. Deformable part models with individual part scaling. In: Proceedings of the British Machine Vision Conference. 2013

[50] GidarisS, Komodakis N. Object detection via a multi-region and semantic segmentation-aware cnn model. In: Proceedings of the IEEE International Conference on Computer Vision. 2015, 1134–1142

[51] GirshickR. Fast r-cnn. 2015, arXiv:1504.08083

[52] GirshickR, Donahue J, DarrellT , MalikJ. Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition, 2014, 580–587

[53] HeK, ZhangX, RenS, Sun J. Spatial pyramid pooling in deep convolutional networks for visual recognition. In: Proceedings of European Conference on Computer Vision. 2014, 346–361

[54] LiangX, LiuS, WeiY, Liu L, LinL , YanS. Computational baby learning. 2014, arXiv:1411.2861

[55] RenS, HeK, GirshickR, Sun J. Faster R-CNN: towards real-time object detection with region proposal networks. 2015, arXiv:1506.01497

[56] RenS, HeK, GirshickR, Zhang X, SunJ . Object detection networks on convolutional feature maps. 2015, arXiv:1504.06066

[57] RenW, HuangK, Tao D, Tan T. Weakly supervised large scale object localization with multiple instance learning and bag splitting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 32(2): 405–416

[58] WanL, EigenD, FergusR. End-to-end integration of a convolutional network, deformable parts model and non-maximum suppression. 2014, arXiv:1411.5309

[59] WangC, HuangK, RenW, Zhang J, MaybankS . Large-scale weakly supervised object localization via latent category learning. IEEE Transactions on Image Processing, 2015, 24(4): 1371–1385

[60] ZhangY, SohnK, VillegasR, Pan G, LeeH . Improving object detection with deep convolutional networks via Bayesian optimization and structured prediction. 2015, arXiv:1504.03293

[61] ZhuY, Urtasun R, SalakhutdinovR , FidlerS. segDeepM: exploiting segmentation and context in deep neural networks for object detection. 2015, arXiv:1502.04275

[62] SongX, WuT, JiaY, Zhu S C. Discriminatively trained and-or tree models for object detection. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2013, 23–28

[63] WangX, LinL, HuangL, Yan S. Incorporating structural alternatives and sharing into hierarchy for multiclass object recognition and detection. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2013, 3334–3341

[64] MarkE, GoolV L, WilliamsC K I , WinnJ, Zisserman A. The PASCAL Visual Object Classes Challenge 2008 (VOC2008) Results. Technical Report. 2008

[65] ChenY, ZhuL, YuilleA. Active mask hierarchies for object detection. In: Proceedings of European Conference on Computer Vision. 2010, 43–56

[66] OttP, Everingham M. Shared parts for deformable part-based models. In: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2011, 1513–1520

[67] ZhangJ, HuangY, HuangK, Wu Z, TanT . Data decomposition and spatial mixture modeling for part based model. In: Proceedings of Asian Conference on Computer Vision. 2012, 123–137

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg