| 1. |
- Danelljan, Martin, 1989-, et al.
(author)
-
ECO: Efficient Convolution Operators for Tracking
- 2017
-
In: Proceedings 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). - : Institute of Electrical and Electronics Engineers (IEEE). - 9781538604571 - 9781538604588 ; , s. 6931-6939
-
Conference paper (peer-reviewed)abstract
- 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.
|
|
| 2. |
- Larsson, Viktor, et al.
(author)
-
Efficient Solvers for Minimal Problems by Syzygy-based Reduction
- 2017
-
In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. - 9781538604588 - 9781538604571 ; , s. 2383-2392
-
Conference paper (peer-reviewed)abstract
- In this paper we study the problem of automatically generatingpolynomial solvers for minimal problems. The maincontribution is a new method for finding small eliminationtemplates by making use of the syzygies (i.e. the polynomialrelations) that exist between the original equations. Usingthese syzygies we can essentially parameterize the setof possible elimination templates.We evaluate our method on a wide variety of problemsfrom geometric computer vision and show improvementcompared to both handcrafted and automatically generatedsolvers. Furthermore we apply our method on two previouslyunsolved relative orientation problems.
|
|
| 3. |
- Butepage, Judith, et al.
(author)
-
Deep representation learning for human motion prediction and classification
- 2017
-
In: 30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017). - : IEEE. - 9781538604571 ; , s. 1591-1599
-
Conference paper (peer-reviewed)abstract
- Generative models of 3D human motion are often restricted to a small number of activities and can therefore not generalize well to novel movements or applications. In this work we propose a deep learning framework for human motion capture data that learns a generic representation from a large corpus of motion capture data and generalizes well to new, unseen, motions. Using an encoding-decoding network that learns to predict future 3D poses from the most recent past, we extract a feature representation of human motion. Most work on deep learning for sequence prediction focuses on video and speech. Since skeletal data has a different structure, we present and evaluate different network architectures that make different assumptions about time dependencies and limb correlations. To quantify the learned features, we use the output of different layers for action classification and visualize the receptive fields of the network units. Our method outperforms the recent state of the art in skeletal motion prediction even though these use action specific training data. Our results show that deep feedforward networks, trained from a generic mocap database, can successfully be used for feature extraction from human motion data and that this representation can be used as a foundation for classification and prediction.
|
|
| 4. |
- Larsson, Viktor, et al.
(author)
-
Compact matrix factorization with dependent subspaces
- 2017
-
In: Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017. - 9781538604571 ; 2017-January, s. 4361-4370
-
Conference paper (peer-reviewed)abstract
- Traditional matrix factorization methods approximate high dimensional data with a low dimensional subspace. This imposes constraints on the matrix elements which allow for estimation of missing entries. A lower rank provides stronger constraints and makes estimation of the missing entries less ambiguous at the cost of measurement fit. In this paper we propose a new factorization model that further constrains the matrix entries. Our approach can be seen as a unification of traditional low-rank matrix factorization and the more recent union-of-subspace approach. It adaptively finds clusters that can be modeled with low dimensional local subspaces and simultaneously uses a global rank constraint to capture the overall scene interactions. For inference we use an energy that penalizes a trade-off between data fit and degrees-of-freedom of the resulting factorization. We show qualitatively and quantitatively that regularizing both local and global dynamics yields significantly improved missing data estimation.
|
|
| 5. |
- Larsson, Viktor, et al.
(author)
-
Compact Matrix Factorization with Dependent Subspaces
- 2017
-
In: 30th IEEE Conference on Computer Vision and Pattern Recognition. - : IEEE. - 1063-6919. - 9781538604571 ; 2017-January, s. 4361-4370
-
Conference paper (peer-reviewed)abstract
- Traditional matrix factorization methods approximate high dimensional data with a low dimensional subspace. This imposes constraints on the matrix elements which allow for estimation of missing entries. A lower rank provides stronger constraints and makes estimation of the missing entries less ambiguous at the cost of measurement fit. In this paper we propose a new factorization model that further constrains the matrix entries. Our approach can be seen as a unification of traditional low-rank matrix factorization and the more recent union-of-subspace approach. It adaptively finds clusters that can be modeled with low dimensional local subspaces and simultaneously uses a global rank constraint to capture the overall scene interactions. For inference we use an energy that penalizes a trade-off between data fit and degrees-of-freedom of the resulting factorization. We show qualitatively and quantitatively that regularizing both local and global dynamics yields significantly improved missing data estimation.
|
|
