Martin Klinkigt

Using a Reference Point for Local Configuration of SIFT-like Features for Object Recognition with Serious Background Clutter

Object recognition can be performed on local or global features. While local features are more robust against occlusions, global features are more powerful to distinguish among many objects. In this paper we propose a novel approach in construction of a shape model from local features aimed at achieving high discriminative power as global features have, while keeping the robustness of local features. We utilize a common reference point expressing the relative position of local features like in a star graph representation. This model is dynamically calculated during recognition which makes it flexible. With our approach we achieve an improved recognition performance of 2% compared to other shape models and even 6% compared to approaches that do not utilize shape information.

Handwriting Reconstruction for a Camera Pen Using Random Dot Patterns

This paper proposes a new method of handwriting reconstruction using a camera pen. We print random dot patterns on the document background to enable retrieval of both the current document and the pen position on this document. Dot arrangements are stored in a hash table using Locally Likely Arrangement Hashing. For retrieval, they are extracted from the camera image and matched to the corresponding points in the hash table. We were able to achieve high retrieval accuracy (81.1~100.0%), given a sufficient amount of visible dots. Using a two-step homography approximation, an accurate image of handwriting can be reconstructed. By using knowledge about document context and a client-server architecture, our method allows real-time processing on ordinary hardware.

From local features to global shape constraints: heterogeneous matching scheme for recognizing objects under serious background clutter

Object recognition in computer vision is the task to categorize images based on their content. With the absence of background clutter in images high recognition performance can be achieved. In this paper we show how the recognition performance is improved even with a high impact of background clutter and without additional information about the image. For this task we segment the image into patches and learn a geometric structure of the object. In evaluations we first show that our system is of comparable performance to other state-of-the-art system and that for a difficult dataset the recognition performance is improved by 13.31%.

Semantic retrieval of images by learning from wikipedia

In this paper we develop a prototype of an image management system supporting the user organizing his images with the help of semantic annotations. The system automatically contributes such annotations to close the gap between images and concepts expressing their content. We propose a novel integration into a Semantic Desktop and the usage of Wikipedia to address the common problem of initial knowledge acquisition and improvement of the performance. With an evaluation on a challenging dataset out system outperforms a purely image processing based approach by 30%.

Generic and specific object recognition for semantic retrieval of images

Since the availability of large digital image collections the need for a proper management of them raises. New technologies as annotations or tagging support the user by doing this task. However, this task is time-consuming and, therefore, automatic annotation systems are requested. Working outside of controlled laboratory environments this request is challenging. In this paper we propose a system automatically adapted to the users needs, providing useful annotations. We utilize Wikipedia to learn instances and abstract classes. With an evaluation in a complex use-case and dataset we show the possibility of such an attempt and achieve practical recognition rates of 26% on specific instance and 64% on abstract class level.