I will present various systems for the recognition of human faces and technical objects. They consist of three steps: feature extraction, solving the correspondence problem, and the actual comparison with stored models. Feature extraction is based on detailed models of cells in the primary visual cortex. The matching will be done by methods based on the Dynamic Link Architecture, which is curre...

With the advent of photo sharing sites like Flickr and Picasa, there has been a growing interest in crowd-sourced modeling and large scale reconstruction using the rich source of Internet imagery. Feature correspondence estimation across multiple images and the global optimization of 3D structure and camera parameters (bundle adjustment) are the primary computational bottlenecks in large scale ...

In this paper a model based approach is presented for generating 3D face models from two stereo images. The proposed method works with a small number of feature points of the face. The automatically registrated point pairs are used to reconstruct their 3D correspondence, then a predefined general face model is adjusted to the reconstructed 3D points. Fitting the general face model to the featur...

Motivation Recently, many researchers have begun to attempt to solve dense correspondence problem for more challenging images which have high variability in terms of photometric and/or geometric conditions [4]. For these challenging scenarios, there exist two principal bottlenecks which make conventional methods provide limited performances; (1) photometric variations derived from different cam...

Detection and learning based appearance feature play the central role in data association based multiple object tracking (MOT), but most recent MOT works usually ignore them and only focus on the hand-crafted feature and association algorithms. In this paper, we explore the high-performance detection and deep learning based appearance feature, and show that they lead to significantly better MOT...

In this contribution we present an algorithm for tracking non-rigid, moving objects in a sequence of colored images, which were recorded by a non-stationary camera. The application background is vision-based driving assistance in the inner city. In an initial step, object parts are determined by a divisive clustering algorithm, which is applied to all pixels in the first image of the sequence. ...

We describe a fast visual feature tracking system which takes advantage of dedicated hardware to perform the computationally intensive step of selection. A software system uses the output of the hardware selector to develop tracks using filtering, data association techniques, and image-based validation.

Common techniques in structure from motion do not explicitly handle foreground occlusions and disocclusions, leading to several trajectories of a single 3D point. Hence, different discontinued trajectories induce a set of (more inaccurate) 3D points instead of a single 3D point, so that it is highly desirable to enforce long continuous trajectories which automatically bridge occlusions after a ...