ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Object detection with constellations of keypoints

Domen Rački (2015) Object detection with constellations of keypoints. MSc thesis.

Download (10Mb)


    Feature-based object detection methods rely on the discriminative nature of features in order to accurately determine the location of a specific object in a test image. From a set of detected features, non-discriminative features are filtered out by means of a similarity threshold, meaning that if a features is very similar to more than one model feature, it is considered to be non-discriminative. However, in cases where an object consists of repeating patterns the similarity threshold proves inefficient since it considers the majority of detected features to be similar to more than one model feature, i.e., non-discriminative. In the context of one-shot learning we propose a constellation model for enhancing basic feature-based object detection methods, with the aim in utilizing the preserved geometry between features to filter out noisy feature matches. This eliminates the need for the similarity threshold. We evaluate the proposed constellation model whit empirically and numerically modelled feature variance and compare it to a baseline feature model. Model evaluation is performed on a challenging real-world dataset, consisting of logotypes in real-world scenarios. We find that the best variation of the constellation model is the model with empirically determined feature variance, which significantly reduces the number of mismatched features, without significantly affecting detection performance.

    Item Type: Thesis (MSc thesis)
    Keywords: one-shot learning, keypoints, geometry, variance, constellation, object detection, SIFT, GHT, MLESAC, MND
    Number of Pages: 81
    Language of Content: English
    Mentor / Comentors:
    Name and SurnameIDFunction
    doc. dr. Matej Kristan4053Mentor
    prof. dr. Thomas PockComentor
    Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=51012&select=(ID=1536313539 )
    Institution: University of Ljubljana
    Department: Faculty of Computer and Information Science
    Item ID: 2984
    Date Deposited: 23 Apr 2015 09:45
    Last Modified: 26 May 2015 12:03
    URI: http://eprints.fri.uni-lj.si/id/eprint/2984

    Actions (login required)

    View Item