Effect of Visual Context Information for Super Resolution Problems

Date

2019-04-26

Author

Akar, Gözde
Aykut, Ekin
Cengiz, Baran
Bocek, Kadircan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

227
views

0
downloads

In this study, the effect of visual context information to the performance of learning-based techniques for the super resolution problem is analyzed. Beside the interpretation of the experimental results in detail, its theoretical reasoning is also achieved in the paper. For the experiments, two different visual datasets composed of natural and remote sensing scenes are utilized. From the experimental results, we observe that keeping visual context information in the course of parameter learning for convolutional neural networks yields better performance compared to the baselines. Moreover, we summarize that fine-tuning pre-trained parameters with the related context yet fewer samples improves the results.

Subject Keywords

Deep learning, Convolutional neural networks, Super resolution

URI

https://hdl.handle.net/11511/47806

DOI

https://doi.org/10.1109/siu.2019.8806598

Collections

Department of Electrical and Electronics Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Theoretical limits and safety considerations for magneto-acousto electrical tomography Ghalichi, Elyar; Gençer, Nevzat Güneri; Department of Electrical and Electronics Engineering (2017) In this study, the performance of Magneto-Acousto-Electrical Tomography (MAET) method is investigated quantitatively by considering interrelations between its sensitivity, resolution and conductivity contrast. An analytical solution for the forward problem of MAET is derived for two-dimensional (2D) concentric bodies by Separation of Variables Method. The electric potential and the acoustic pressure are separated to their angular and radial components. The series coefficients for these solutions are obtaine...
Theoretical limits to sensitivity and resolution in magneto-acousto-electrical tomography GHALICHI, Elyar; Gençer, Nevzat Güneri (2017-10-21) In this study, the performance of magneto-acousto-electrical tomography (MAET) is investigated quantitatively by considering interrelations between its sensitivity, resolution and conductivity contrast. An analytical solution for the forward problem of MAET is derived for two-dimensional (2D) concentric bodies by the separation of variables method. The electric potential and the acoustic pressure are separated into their angular and radial components. The series coefficients for these solutions are obtained...
Effects of modulation type on blind SIMO commmunications systems Karakutuk, Serkan; Tuncer, Temel Engin (2007-06-13) In frame based communication systems, blind identification of frame length and offset from limited number of observations has an important affect on the performance of blind channel identification and equalization algorithms. In [1], for guard interval transmission schemes, a novel blind frame length and offset estimation method is proposed. In this paper, the performance of methods proposed in [1] for fixed frame length transmission are studied.
Effect of human prior knowledge on game success and comparison with reinforcement learning Hasanoğlu, Mert.; Çakır, Murat Perit; Department of Cognitive Sciences (2019) This study aims to find out the effect of prior knowledge on the success of humans in a non-rewarding game environment, and then to compare human performance with a reinforcement learning method in an effort to observe to what extent this method can be brought closer to human behavior and performance with the data obtained. For this purpose, different versions of a simple 2D game were used, and data were collected from 32 participants. At the end of the experiment, it is concluded that prior knowledge, such...
Transfer of Learning on a Spatial Memory Task between the Blind and Sighted People AKPINAR, SELÇUK; Popovic, Stevo; Kirazcı, Sadettin (2012-12-01) The purpose of this study was to analyze the effect of two different types of feedback on a spatial memory task between the blind and blindfolded-sighted participants. Participants tried to estimate the predetermined distance by using their dominant hands. Both blind and blindfolded-sighted groups were randomly divided into two feedback subgroups as "100% frequency" and "10% bandwidth". The score of the participants was given verbally to the participants as knowledge of results (KR). The target distance was...

Citation Formats

G. Akar, E. Aykut, B. Cengiz, and K. Bocek, “Effect of Visual Context Information for Super Resolution Problems,” 2019, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/47806.