In a number of practical scenarios, such as video conferencing and visual human/computer interaction, objects that belong to a well defined class are segmented, normalized, and encoded, after which they are stored and/or transmitted, and subsequently reconstructed. The Karhunen-Loève Transform (KLT) optimally concentrates the signal power in a relatively small number of uncorrelated coefficient...

This Master thesis deals with an approach for 2D-3D pose estimation, which relies on texture information on the surface mesh of an object model. The textured surface mesh is projected in a virtual image, and a block matching algorithm is employed to determine correspondences between midpoints of surface patches to the image data. To facilitate this purpose, the block matching algorithm needs to...

Three-dimensional head pose estimation from a single 2D image is a challenging task with extensive applications. Existing approaches lack the capability to deal with multiple pose-related and -unrelated factors in a uniform way. Most of them can provide only one-dimensional yaw estimation and suffer from limited representation ability for out-of-sample testing inputs. These drawbacks lead to li...

In the image parameter estimation by the linear regression, it has very high degrees of freedom for the decision of regression coefficients, because the dimension of image vector is huge high. In this paper, we discuss its potential by the learning of the dense samples. For the learning process, we employed a sequential regression coefficient calculation algorithm and realize its calculation fo...

We propose an approach to 3D facial pose estimation that, unlike most state-of-the-art techniques, can handle arbitrary poses, including extreme out-of-plane rotations, background clutter and facial expressions. It relies on a large database of registered face images of different people viewed from several perspectives. We use a powerful image retrieval technique to match the input image agains...

We present a fast and efficient approach for joint person detection pose estimation optimized automated driving (AD) in urban scenarios. use multitask weight sharing architecture to jointly train estimation. This modular allows us accommodate different downstream tasks the future. By systematic large-scale experiments on Tsinghua-Daimler Urban Pose Dataset (TDUP), we obtain multiple models with...

We propose an approach to multi-view object detection and pose estimation that considers combinations of single-view estimates. It can be used with most existing single-view pose estimation systems, and can produce improved results even if the individual pose estimates are incoherent. The method is introduced in the context of an existing, probabilistic, view-based detection and pose estimation...

This paper presents a local image feature, based on the logpolar transform which renders it invariant to orientation and scale variations. It is shown that this feature can be used for pose estimation of 3D objects with unknown pose, with cluttered background and with occlusion. The proposed method is compared to a previously published one and the new feature is found to be about as good or bet...

In this paper, we present a two-step methodology to improve existing human pose estimation methods from a single depth image. Instead of learning the direct mapping from the depth image to the 3D pose, we first estimate the orientation of the standing person seen by the camera and then use this information to dynamically select a pose estimation model suited for this particular orientation. We ...