Computer Vision Systems: 10th International Conference, ICVS by Lazaros Nalpantidis, Volker Krüger, Jan-Olof Eklundh,

By Lazaros Nalpantidis, Volker Krüger, Jan-Olof Eklundh, Antonios Gasteratos

This e-book constitutes the refereed complaints of the tenth foreign convention on laptop imaginative and prescient structures, ICVS 2015, held in Copenhagen, Denmark, in July 2015. The forty eight papers provided have been rigorously reviewed and chosen from ninety two submissions. The paper are prepared in topical sections on organic and cognitive imaginative and prescient; hardware-implemented and real-time imaginative and prescient structures; high-level imaginative and prescient; studying and variation; robotic imaginative and prescient; and imaginative and prescient platforms applications.

Show description

Read or Download Computer Vision Systems: 10th International Conference, ICVS 2015, Copenhagen, Denmark, July 6-9, 2015, Proceedings PDF

Similar international_1 books

Energy Minimization Methods in Computer Vision and Pattern Recognition: Third International Workshop, EMMCVPR 2001 Sophia Antipolis, France, September 3–5, 2001 Proceedings

This quantity involves the forty two papers awarded on the overseas Workshop on power Minimization tools in computing device imaginative and prescient and development acceptance (EMMCVPR2001),whichwasheldatINRIA(InstitutNationaldeRechercheen Informatique et en Automatique) in Sophia Antipolis, France, from September three via September five, 2001.

The Proceedings of the Second International Conference on Communications, Signal Processing, and Systems

The court cases of the second one foreign convention on Communications, sign Processing, and structures presents the state-of-art advancements of Communications, sign Processing, and structures. The convention coated such themes as instant communications, networks, structures, sign processing for communications.

Additional resources for Computer Vision Systems: 10th International Conference, ICVS 2015, Copenhagen, Denmark, July 6-9, 2015, Proceedings

Sample text

37 %). Moreover, when the cell activiy is thresholded the classification accuracy is improved. 19 %. 78 % of correct classifications. 4 Conclusions The goal of this work was to implement a bio-inspired descriptor, mimicking some functionalities of the visual system. From biology, it is well known that the retina extracts details from images using a Difference of Gaussians (DoG) of different sizes and encodes such differences with action potentials. We have presented a set of modifications to FREAK which are more biologically inspired.

70 In this table we can observe that eliminating the less discriminative pairs from the descriptor increases the performance. Best results are obtained when the size is reduced to 256 bytes. e. 64 bytes). In addition to this, in all our experiments the scale and orientation normalization is not used, since we are using a dense grid and not a keypoint detector as in the original idea. The confusion matrix from one run of the DescrC S ON OF F T h descriptor is shown in Fig. 3, where row names are true labels and column names are the predicted ones.

Intell. 32(5), 815–830 (2010) 11. : Dense interest points. In: The Twenty-Third IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2010, pp. 2281–2288, San Francisco, CA, USA, 13–18 June 2010 (2010) 12. : VLFeat: an open and portable library of computer vision algorithms (2008). org/ 13. : Improving texture categorization with biologically inspired filtering. Image Vis. Comput. 32, 424–436 (2013) 14. : CS-FREAK: an improved binary descriptor. , Huang, K. ) IGTA 2014. CCIS, vol. 437, pp.

Download PDF sample

Rated 4.33 of 5 – based on 17 votes