Elliptic shape prior for object 2D-3D pose estimation using circular feature

Many objects in real world have circular feature. It is a difficult task to obtain the 2D-3D pose estimation using circular feature in challenging scenarios. This paper proposes a method to incorporate elliptic shape prior for object pose estimation using a level set method. The relationship between the projection of the circular feature of a 3D object and the signed distance function corresponding to it is analyzed to yield a 2D elliptic shape prior. The method employs the combination of the grayscale histogram, the intensities of edge, and the smoothness distribution as main image feature descriptors that define the image statistical measure model. The elliptic shape prior combined with the image statistical measure energy model drives the elliptic shape contour to the projection of the circular feature of the 3D object with the current pose into the image plane. These works effectively reduce the impacts of the challenging scenarios on the pose estimate results. In addition, the method utilizes particle filters that take into account the motion dynamics of the object among scene frames, and this work provides the robust method for object 2D-3D pose estimation using circular feature in a challenging environment. Various numerical experiments are illustrated to show the performance and advantages of the proposed method.

[1]  Siva Ram Krishna Vadali,et al.  Reliable pose estimation of underwater dock using single camera: a scene invariant approach , 2015, Machine Vision and Applications.

[2]  Yun Fu,et al.  Arc-Support Line Segments Revisited: An Efficient High-Quality Ellipse Detection , 2018, IEEE Transactions on Image Processing.

[3]  Cai Meng,et al.  Monocular pose measurement method based on circle and line features , 2016 .

[4]  David Zhang,et al.  Joint Registration and Active Contour Segmentation for Object Tracking , 2013, IEEE Transactions on Circuits and Systems for Video Technology.

[5]  Robin J. Evans,et al.  Fundamentals of Object Tracking , 2011 .

[6]  Meng Li,et al.  Direct solution for pose estimation of single circle with detected centre , 2016 .

[7]  Roland Glowinski,et al.  Convex Shape Prior for Multi-Object Segmentation Using a Single Level Set Function , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[8]  Vitaly Kober,et al.  Accurate three-dimensional pose recognition from monocular images using template matched filtering , 2016 .

[9]  Anthony J. Yezzi,et al.  A Geometric Approach to Joint 2D Region-Based Segmentation and 3D Pose Estimation Using a 3D Shape Prior , 2010, SIAM J. Imaging Sci..

[10]  Derong Chen,et al.  Object tracking using both a kernel and a non-parametric active contour model , 2018, Neurocomputing.

[11]  Vitaly Kober,et al.  Three-dimensional pose tracking by image correlation and particle filtering , 2018 .

[12]  Bodo Rosenhahn,et al.  Three-Dimensional Shape Knowledge for Joint Image Segmentation and Pose Estimation , 2005, DAGM-Symposium.

[13]  Yiu Cheung Shiu,et al.  3D location of circular and spherical features by monocular model-based vision , 1989, Conference Proceedings., IEEE International Conference on Systems, Man and Cybernetics.

[14]  Bin Huang,et al.  General fusion frame of circles and points in vision pose estimation , 2018 .

[15]  Vitaly Kober,et al.  Target tracking in nonuniform illumination conditions using locally adaptive correlation filters , 2014 .

[16]  Allen R. Tannenbaum,et al.  Particle filters and occlusion handling for rigid 2D-3D pose tracking , 2013, Comput. Vis. Image Underst..

[17]  Jianye Liu,et al.  Vision pose estimation from planar dual circles in a single image , 2016 .