A Particle Filter Based Probabilistic Fusion Framework for Simultaneous Recognition and Pose Estimation of 3 D Objects in a Sequence of Images

We describe a framework for robust object recognition and pose estimation of 3D object which is using a sequence of images and probabilistic method. Using a sequence of images in multiple views has great advantage for robust object recognition and pose estimation in noisy and ill-conditioned environment texture, occlusion, illuminate and camera pose. For recognizing an object and estimating its pose, we present it as a particle filter based probabilistic method with information of a sequence of images. It means that object pose represents probability distribution by particles in 3D space, and updated particles by consecutive observation in a sequence of images are converged to a single particle. The proposed method allows an easy integration of multiple evidences such photometric and geometric features as SIFT, color, 3D line, 2D square, and etc. The experimental results with stereo camera show the validity of the proposed method in an environment containing both textured and texture-less objects.