Anil Ozdemir
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Anil Ozdemir has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Andrew B. Barron
School of Natural Sciences, Macquarie University
1 shared publication - Andrew Philippides
AI Group, Department of Informatics, University of Sussex, Brighton BN1 9RH, UK
1 shared publication Ayca Cinaroglu
Department of Molecular Biology and Genetics, Bogazici University, Istanbul, Turkey
1 shared publicationCeren Ergorul
Department of Molecular Biology and Genetics, Bogazici University, Istanbul, Turkey
1 shared publication- David Hicks
Centre National de la Recherche Scientifique, Université de Strasbourg, Institut des Neurosciences Cellulaires et Intégratives, Strasbourg, France
1 shared publication - Eleni Vasilaki
University of Sheffield
1 shared publication
Latest Publications
- 11 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 7 (2) , 4520-4527
EchoVPR: Echo State Networks for Visual Place Recognition
Recognising previously visited locations is an important, but unsolved, task in autonomous navigation. Current visual place recognition (VPR) benchmarks typically challenge models to recover the position of a query image (or images) from sequential datasets that include both spatial and temporal components. Recently, Echo State Network (ESN) varieties have proven particularly powerful at solving machine learning tasks that require spatio-temporal modelling. These networks are simple, yet powerful neural architectures that—exhibiting memory over multiple time-scales and non-linear high-dimensional representations—can discover temporal relations in the data while still maintaining linearity in the learning time. In this letter, we present a series of ESNs and analyse their applicability to the VPR problem. We report that the addition of ESNs to pre-processed convolutional neural networks led to a dramatic boost in performance in comparison to non-recurrent networks in five out of six standard benchmarks (GardensPoint, SPEDTest, ESSEX3IN1, Oxford RobotCar, and Nordland), demonstrating that ESNs are able to capture the temporal structure inherent in VPR problems. Moreover, we show that models that include ESNs can outperform class-leading VPR models which also exploit the sequential dynamics of the data. Finally, our results demonstrate that ESNs improve generalisation abilities, robustness, and accuracy further supporting their suitability to VPR applications.
- 11 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 7 (2) , 4520-4527
EchoVPR: Echo State Networks for Visual Place Recognition
Recognising previously visited locations is an important, but unsolved, task in autonomous navigation. Current visual place recognition (VPR) benchmarks typically challenge models to recover the position of a query image (or images) from sequential datasets that include both spatial and temporal components. Recently, Echo State Network (ESN) varieties have proven particularly powerful at solving machine learning tasks that require spatio-temporal modelling. These networks are simple, yet powerful neural architectures that—exhibiting memory over multiple time-scales and non-linear high-dimensional representations—can discover temporal relations in the data while still maintaining linearity in the learning time. In this letter, we present a series of ESNs and analyse their applicability to the VPR problem. We report that the addition of ESNs to pre-processed convolutional neural networks led to a dramatic boost in performance in comparison to non-recurrent networks in five out of six standard benchmarks (GardensPoint, SPEDTest, ESSEX3IN1, Oxford RobotCar, and Nordland), demonstrating that ESNs are able to capture the temporal structure inherent in VPR problems. Moreover, we show that models that include ESNs can outperform class-leading VPR models which also exploit the sequential dynamics of the data. Finally, our results demonstrate that ESNs improve generalisation abilities, robustness, and accuracy further supporting their suitability to VPR applications.
- 21 December 2004
- Published by Wiley in Journal of Neuroscience Research
- Vol. 79 (3) , 329-339
Expression and possible function of fibroblast growth factor 9 (FGF9) and its cognate receptors FGFR2 and FGFR3 in postnatal and adult retina
Fibroblast growth factors (FGFs) are important regulators of retinal development and survival. We examined the expression and distribution of FGF9 and its preferred receptors FGFR2IIIc and FGFR3IIIc in this tissue. FGF9 transcripts in whole rat retina were detected by RT‐PCR but were not present in purified cultured Müller glia. Transcripts appeared as 3.2‐kb and 4.0‐kb bands on Northern blots, and Western blotting of whole retina revealed FGF9‐immunoreactive bands at 30 and 55 kDa. FGF9 mRNA demonstrated a biphasic expression profile, elevated at birth and adulthood, but relatively decreased during terminal retinal differentiation (4–14 days postnatal). Antibody labeling broadly reflected these findings: staining in vivo was observed mainly in the inner retina (and outer plexiform layer in adults) whereas FGF9 was not detectable in cultured Müller glia. In adults, FGF9 in situ hybridization also showed a detectable signal in inner retina. FGFR2IIIc and FGFR3IIIc were detected by RT‐PCR, and Western blotting showed both FGFRs existed as multiple forms between ∼100–200 kDa. FGFR2 and FGFR3 antibodies showed prominent labeling in the inner retina, especially in proliferating cultured Müller glia. Exogenous FGF9 elicited a dose‐dependent increase in Müller glial proliferation in vitro. These data suggest a role for FGF9 in retinal differentiation and maturation, possibly representing a neuronally derived factor acting upon glial (and other) cells.
- 21 December 2004
- Published by Wiley in Journal of Neuroscience Research
- Vol. 79 (3) , 329-339
Expression and possible function of fibroblast growth factor 9 (FGF9) and its cognate receptors FGFR2 and FGFR3 in postnatal and adult retina
Fibroblast growth factors (FGFs) are important regulators of retinal development and survival. We examined the expression and distribution of FGF9 and its preferred receptors FGFR2IIIc and FGFR3IIIc in this tissue. FGF9 transcripts in whole rat retina were detected by RT‐PCR but were not present in purified cultured Müller glia. Transcripts appeared as 3.2‐kb and 4.0‐kb bands on Northern blots, and Western blotting of whole retina revealed FGF9‐immunoreactive bands at 30 and 55 kDa. FGF9 mRNA demonstrated a biphasic expression profile, elevated at birth and adulthood, but relatively decreased during terminal retinal differentiation (4–14 days postnatal). Antibody labeling broadly reflected these findings: staining in vivo was observed mainly in the inner retina (and outer plexiform layer in adults) whereas FGF9 was not detectable in cultured Müller glia. In adults, FGF9 in situ hybridization also showed a detectable signal in inner retina. FGFR2IIIc and FGFR3IIIc were detected by RT‐PCR, and Western blotting showed both FGFRs existed as multiple forms between ∼100–200 kDa. FGFR2 and FGFR3 antibodies showed prominent labeling in the inner retina, especially in proliferating cultured Müller glia. Exogenous FGF9 elicited a dose‐dependent increase in Müller glial proliferation in vitro. These data suggest a role for FGF9 in retinal differentiation and maturation, possibly representing a neuronally derived factor acting upon glial (and other) cells.