Computer Engineering / Bilgisayar Mühendisliği

Permanent URI for this collectionhttps://hdl.handle.net/11147/10

Browse

Filters

2012 - 201922020 - 20222

Settings

Author: search.filters.author.Baştanlar, YalınScopus Q: search.filters.scopusQuartile.Q1

Search Results

Now showing 1 - 4 of 4
  • Conference Object
    Citation - WoS: 7
    Citation - Scopus: 6
    Semantic Pose Verification for Outdoor Visual Localization With Self-Supervised Contrastive Learning
    (IEEE, 2022) Guerrero, Jose J.; Orhan, Semih; Baştanlar, Yalın
    Any city-scale visual localization system has to overcome long-term appearance changes, such as varying illumination conditions or seasonal changes between query and database images. Since semantic content is more robust to such changes, we exploit semantic information to improve visual localization. In our scenario, the database consists of gnomonic views generated from panoramic images (e.g. Google Street View) and query images are collected with a standard field-of-view camera at a different time. To improve localization, we check the semantic similarity between query and database images, which is not trivial since the position and viewpoint of the cameras do not exactly match. To learn similarity, we propose training a CNN in a self-supervised fashion with contrastive learning on a dataset of semantically segmented images. With experiments we showed that this semantic similarity estimation approach works better than measuring the similarity at pixel-level. Finally, we used the semantic similarity scores to verify the retrievals obtained by a state-of-the-art visual localization method and observed that contrastive learning-based pose verification increases top-1 recall value to 0.90 which corresponds to a 2% improvement.
  • Article
    Citation - WoS: 7
    Citation - Scopus: 8
    Long-Term Image-Based Vehicle Localization Improved With Learnt Semantic Descriptors
    (Elsevier, 2022) Çınaroğlu, İbrahim; Baştanlar, Yalın
    Vision based solutions for the localization of vehicles have become popular recently. In this study, we employ an image retrieval based visual localization approach, in which database images are kept with GPS coordinates and the location of the retrieved database image serves as the position estimate of the query image in a city scale driving scenario. Regarding this approach, most existing studies only use descriptors extracted from RGB images and do not exploit semantic content. We show that localization can be improved via descriptors extracted from semantically segmented images, especially when the environment is subjected to severe illumination, seasonal or other long-term changes. We worked on two separate visual localization datasets, one of which (Malaga Streetview Challenge) has been generated by us and made publicly available. Following the extraction of semantic labels in images, we trained a CNN model for localization in a weakly-supervised fashion with triplet ranking loss. The optimized semantic descriptor can be used on its own for localization or preferably it can be used together with a state-of-the-art RGB image based descriptor in hybrid fashion to improve accuracy. Our experiments reveal that the proposed hybrid method is able to increase the localization performance of the standard (RGB image based) approach up to 7.7% regarding Top-1 Recall values.
  • Article
    Citation - WoS: 11
    Citation - Scopus: 15
    A Simplified Two-View Geometry Based External Calibration Method for Omnidirectional and Ptz Camera Pairs
    (Elsevier Ltd., 2016) Baştanlar, Yalın
    The external calibration of a camera system is essential for most of the applications that involve an omnidirectional and a pan-tilt-zoom (PTZ) camera. The methods in the literature fall into two major categories; (1) a complete external calibration of the system which allows all degrees of freedom but highly time consuming, (2) spatial mapping between the pixel coordinates in omnidirectional camera and pan/tilt angles of the PTZ camera instead of explicitly computing the rotation and translation. Most methods in this category make restrictive assumptions about the camera setup such as optical axes of the cameras coincide. We propose an external calibration method that is effective and practical. Using the two-view geometry principles and making reasonable assumptions about the camera setup, calibration is performed with just two scene points. We extract rotation using the point correspondences in images. Locating the PTZ camera in the omnidirectional image is used to find the translation parameters and the real distance between the two scene points lets us compute the translation in correct scale. Results of the simulated and real image experiments show that our method works effectively in real world cases and its accuracy is comparable to the state-of-the-art methods.
  • Article
    Citation - WoS: 69
    Citation - Scopus: 89
    Joint Optimization for Object Class Segmentation and Dense Stereo Reconstruction
    (Springer Verlag, 2012) Ladicky, Lubor; Sturgess, Paul; Russell, Chris; Sengupta, Sunando; Baştanlar, Yalın; Clocksin, William; Torr, Philip H.S.
    The problems of dense stereo reconstruction and object class segmentation can both be formulated as Random Field labeling problems, in which every pixel in the image is assigned a label corresponding to either its disparity, or an object class such as road or building. While these two problems are mutually informative, no attempt has been made to jointly optimize their labelings. In this work we provide a flexible framework configured via cross-validation that unifies the two problems and demonstrate that, by resolving ambiguities, which would be present in real world data if the two problems were considered separately, joint optimization of the two problems substantially improves performance. To evaluate our method, we augment the Leu-ven data set (http://cms.brookes.ac.uk/research/visiongroup/ files/Leuven.zip), which is a stereo video shot from a car driving around the streets of Leuven, with 70 hand labeled object class and disparity maps. We hope that the release of these annotations will stimulate further work in the challenging domain of street-view analysis. Complete source code is publicly available (http://cms.brookes.ac.uk/ staff/Philip-Torr/ale.htm). © 2011 Springer Science+Business Media, LLC.