Good Features to Correlate for Visual Tracking

Download

index.pdf

Date

2018-05-01

Author

Gundogdu, Erhan
Alatan, Abdullah Aydın

Metadata

Show full item record

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

Item Usage Stats

144
views

0
downloads

During the recent years, correlation filters have shown dominant and spectacular results for visual object tracking. The types of the features that are employed in this family of trackers significantly affect the performance of visual tracking. The ultimate goal is to utilize the robust features invariant to any kind of appearance change of the object, while predicting the object location as properly as in the case of no appearance change. As the deep learning based methods have emerged, the study of learning features for specific tasks has accelerated. For instance, discriminative visual tracking methods based on deep architectures have been studied with promising performance. Nevertheless, correlation filter based (CFB) trackers confine themselves to use the pre-trained networks, which are trained for object classification problem. To this end, in this manuscript the problem of learning deep fully convolutional features for the CFB visual tracking is formulated. In order to learn the proposed model, a novel and efficient backpropagation algorithm is presented based on the loss function of the network. The proposed learning framework enables the network model to be flexible for a custom design. Moreover, it alleviates the dependency on the network trained for classification. Extensive performance analysis shows the efficacy of the proposed custom design in the CFB tracking framework. By fine-tuning the convolutional parts of a state-of-the-art network and integrating this model to a CFB tracker, which is the top performing one of VOT2016, 18% increase is achieved in terms of expected average overlap, and tracking failures are decreased by 25%, while maintaining the superiority over the state-of-the-art methods in OTB-2013 and OTB-2015 tracking datasets.

Subject Keywords

Software, Computer Graphics and Computer-Aided Design

URI

https://hdl.handle.net/11511/38340

Journal

IEEE TRANSACTIONS ON IMAGE PROCESSING

DOI

https://doi.org/10.1109/tip.2018.2806280

Collections

Department of Electrical and Electronics Engineering, Article

Suggestions

OpenMETU
Core

Extending Correlation Filter-Based Visual Tracking by Tree-Structured Ensemble and Spatial Windowing Gundogdu, Erhan; Özkan, Huseyin; Alatan, Abdullah Aydın (Institute of Electrical and Electronics Engineers (IEEE), 2017-11-01) Correlation filters have been successfully used in visual tracking due to their modeling power and computational efficiency. However, the state-of-the-art correlation filter-based (CFB) tracking algorithms tend to quickly discard the previous poses of the target, since they consider only a single filter in their models. On the contrary, our approach is to register multiple CFB trackers for previous poses and exploit the registered knowledge when an appearance change occurs. To this end, we propose a novel t...
Fuzzy spatial data cube construction and its use in association rule mining Işık, Narin; Yazıcı, Adnan; Department of Computer Engineering (2005) The popularity of spatial databases increases since the amount of the spatial data that need to be handled has increased by the use of digital maps, images from satellites, video cameras, medical equipment, sensor networks, etc. Spatial data are difficult to examine and extract interesting knowledge; hence, applications that assist decision-making about spatial data like weather forecasting, traffic supervision, mobile communication, etc. have been introduced. In this thesis, more natural and precise knowle...
Geometry-Aware Neighborhood Search for Learning Local Models for Image Superresolution Ferreira, Julio Cesar; Vural, Elif; Guillemot, Christine (Institute of Electrical and Electronics Engineers (IEEE), 2016-03-01) Local learning of sparse image models has proved to be very effective to solve inverse problems in many computer vision applications. To learn such models, the data samples are often clustered using the K-means algorithm with the Euclidean distance as a dissimilarity metric. However, the Euclidean distance may not always be a good dissimilarity measure for comparing data samples lying on a manifold. In this paper, we propose two algorithms for determining a local subset of training samples from which a good...
Deep Joint Deinterlacing and Denoising for Single Shot Dual-ISO HDR Reconstruction Cogalan, Ugur; Akyüz, Ahmet Oğuz (Institute of Electrical and Electronics Engineers (IEEE), 2020-01-01) HDR images have traditionally been obtained by merging multiple exposures each captured with a different exposure time. However, this approach entails longer capture times and necessitates deghosting if the captured scene contains moving objects. With the advent of modern camera sensors that can perform per-pixel exposure modulation, it is now possible to capture all of the required exposures within a single shot. The new challenge then becomes how to best combine different pixels with different exposure va...
Improving interactive classification of satellite image content Tekkaya, Gökhan; Atalay, Mehmet Volkan; Department of Computer Engineering (2007) Interactive classication is an attractive alternative and complementary for automatic classication of satellite image content, since the subject is visual and there are not yet powerful computational features corresponding to the sought visual features. In this study, we improve our previous attempt by building a more stable software system with better capabilities for interactive classication of the content of satellite images. The system allows user to indicate a few number of image regions that contain a...

Citation Formats

E. Gundogdu and A. A. Alatan, “Good Features to Correlate for Visual Tracking,” IEEE TRANSACTIONS ON IMAGE PROCESSING, pp. 2526–2540, 2018, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/38340.