Search: id:"swepub:oai:DiVA.org:liu-144284" >
ECO: Efficient Conv...
ECO: Efficient Convolution Operators for Tracking
-
- Danelljan, Martin, 1989- (author)
- Linköpings universitet,Datorseende,Tekniska fakulteten
-
- Bhat, Goutam (author)
- Linköpings universitet,Datorseende,Tekniska fakulteten
-
- Khan, Fahad Shahbaz, 1983- (author)
- Linköpings universitet,Datorseende,Tekniska fakulteten
-
show more...
-
- Felsberg, Michael, 1974- (author)
- Linköpings universitet,Datorseende,Tekniska fakulteten
-
show less...
-
(creator_code:org_t)
- Institute of Electrical and Electronics Engineers (IEEE), 2017
- 2017
- English.
-
In: Proceedings 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). - : Institute of Electrical and Electronics Engineers (IEEE). - 9781538604571 - 9781538604588 ; , s. 6931-6939
- Related links:
-
https://liu.diva-por... (primary) (Raw object)
-
show more...
-
https://urn.kb.se/re...
-
https://doi.org/10.1...
-
show less...
Abstract
Subject headings
Close
- In recent years, Discriminative Correlation Filter (DCF) based methods have significantly advanced the state-of-the-art in tracking. However, in the pursuit of ever increasing tracking performance, their characteristic speed and real-time capability have gradually faded. Further, the increasingly complex models, with massive number of trainable parameters, have introduced the risk of severe over-fitting. In this work, we tackle the key causes behind the problems of computational complexity and over-fitting, with the aim of simultaneously improving both speed and performance. We revisit the core DCF formulation and introduce: (i) a factorized convolution operator, which drastically reduces the number of parameters in the model; (ii) a compact generative model of the training sample distribution, that significantly reduces memory and time complexity, while providing better diversity of samples; (iii) a conservative model update strategy with improved robustness and reduced complexity. We perform comprehensive experiments on four benchmarks: VOT2016, UAV123, OTB-2015, and Temple-Color. When using expensive deep features, our tracker provides a 20-fold speedup and achieves a 13.0% relative gain in Expected Average Overlap compared to the top ranked method [12] in the VOT2016 challenge. Moreover, our fast variant, using hand-crafted features, operates at 60 Hz on a single CPU, while obtaining 65.0% AUC on OTB-2015.
Subject headings
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
Publication and Content Type
- ref (subject category)
- kon (subject category)
Find in a library
To the university's database