Abstract
Object detection is an important research field in computer vision, previous studies have focused on detection in standard RGB images, with the popularity of motion-sensing camera, such as Kinect, combining RGB and depth data in object detection becomes a new challenge. In this paper, the object model is analyzed thoroughly with the trait of depth information, a criteria CIR (Contour Intensity Ratio) is given out to measure the contour strength of model, and then the advantage of depth data in encoding the contour strength is illustrated. The Deformable Part Model (DPM) [1] and Histogram of Oriented Depths (HOD) [2] are combined to a new RGB-D joint detection strategy, and the 4 rules of depth filtration after detection as well as part-class depth filtration are put forward, the important role of depth information for object detection is confirmed by human detection experiment.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Felzenszwalb, P., McAllester, D., Ramanan, D.: A discriminatively trained, multiscale, deformable part model. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2008, pp. 1–8. IEEE (2008)
Spinello, L., Arras, K.O.: People detection in rgb-d data. In: 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 3838–3843. IEEE (2011)
Wu, J., Geyer, C., Rehg, J.M.: Real-time human detection using contour cues. In: 2011 IEEE International Conference on Robotics and Automation (ICRA), pp. 860–867. IEEE (2011)
Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, vol. 1, pp. 886–893. IEEE (2005)
Felzenszwalb, P.F., Girshick, R.B., McAllester, D., Ramanan, D.: Object detection with discriminatively trained part-based models. IEEE Transactions on Pattern Analysis and Machine Intelligence 32, 1627–1645 (2010)
Borràs, R., Lapedriza, À., Igual, L.: Depth information in human gait analysis: An experimental study on gender recognition. In: Campilho, A., Kamel, M. (eds.) ICIAR 2012, Part II. LNCS, vol. 7325, pp. 98–105. Springer, Heidelberg (2012)
Wang, N., Gong, X., Liu, J.: A new depth descriptor for pedestrian detection in rgb-d images. In: 2012 21st International Conference on Pattern Recognition (ICPR), pp. 3688–3691. IEEE (2012)
Microsoft: Xbox 360 Kinect Sensor Manual (2010)
Shao, L., Han, J., Xu, D., Shotton, J.: Computer vision for rgb-d sensors: Kinect and its applications. IEEE Transactions on Cybernetics 43, 1314–1317 (2013)
Everingham, M., van Gool, L., Williams, C.K.I., Zisserman, A.: The pascal visual object classes challenge 2006 (voc 2006) results (2006), http://www.pascalnetwork.org/challenges/VOC/voc2006/
Everingham, M., van Gool, L., Williams, C.K.I., Zisserman, A.: The pascal visual object classes challenge 2007 (voc 2007) results (2007), http://www.pascalnetwork.org/challenges/VOC/voc2007/
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Li, C., Ma, S., Wang, T., Sheng, H., Xiong, Z. (2014). Object Detection Using Deformable Part Model in RGB-D Data. In: Bebis, G., et al. Advances in Visual Computing. ISVC 2014. Lecture Notes in Computer Science, vol 8887. Springer, Cham. https://doi.org/10.1007/978-3-319-14249-4_65
Download citation
DOI: https://doi.org/10.1007/978-3-319-14249-4_65
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-14248-7
Online ISBN: 978-3-319-14249-4
eBook Packages: Computer ScienceComputer Science (R0)