2015
DOI: 10.1007/978-3-319-23437-3_30
|View full text |Cite
|
Sign up to set email alerts
|

Towards the Automatic Definition of the Objective Function for Model-Based 3D Hand Tracking

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2017
2017
2019
2019

Publication Types

Select...
2
1

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(3 citation statements)
references
References 13 publications
0
3
0
Order By: Relevance
“…For these approaches, since the learned cost function is differentiable w.r.t the parameters, they use Gauss-Newton method, similar to that in Lucas-Kanade [5], to optimize the learned cost for the unseen data. In [80], Paliouras and Argyros used nonlinear regressors and random forests with various features to learn the cost function for hand pose estimation in RGBD images, where the learned cost function was optimized using particle swarm optimization.…”
Section: Learning Cost Functionsmentioning
confidence: 99%
“…For these approaches, since the learned cost function is differentiable w.r.t the parameters, they use Gauss-Newton method, similar to that in Lucas-Kanade [5], to optimize the learned cost for the unseen data. In [80], Paliouras and Argyros used nonlinear regressors and random forests with various features to learn the cost function for hand pose estimation in RGBD images, where the learned cost function was optimized using particle swarm optimization.…”
Section: Learning Cost Functionsmentioning
confidence: 99%
“…Instead of manually design a cost function, several works proposed to use machine learning techniques to learn a cost function from available training data. For example, kernel SVM [10], boosting [11], metric learning [12], and nonlinear regressors [13] have been used to learn a cost function for image-based tracking, alignment, and pose estimation. Once a cost function is learned, the optimal parameters are solved using search algorithms such as descent methods or particle swarm optimization.…”
Section: Optimization In Computer Visionmentioning
confidence: 99%
“…Regarding the formulation of the objective function, an idea worth investigating as an alternative to compare occupancy maps such as the skin color maps, is the Jaccard distance [82]. Furthermore, apart from designing the objective function by hand, it is possible to automate the process, yielding potentially more accurate results [77].…”
Section: Future Workmentioning
confidence: 99%