Estimating Human Pose from Occluded Images

Huang, Jia-Bin; Yang, Ming–Hsuan

doi:10.1007/978-3-642-12307-8_5

Cited by 21 publications

(13 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Improved versions include an optimized objective, like a parts objective (“PARTS”) based on discrete information gain [9], while other works report the generalization problem of the specified objective [270,271]. Furthermore, sparse representation (SR) is used to extract the most significant training samples, and later on, all estimations are carried out based on these samples [168,169,170,171]. …”

Section: Methodologiesmentioning

confidence: 99%

Human Pose Estimation from Monocular Images: A Comprehensive Survey

Gong

Zhang

Gonzàlez

et al. 2016

Sensors

130

View full text Add to dashboard Cite

Human pose estimation refers to the estimation of the location of body parts and how they are connected in an image. Human pose estimation from monocular images has wide applications (e.g., image indexing). Several surveys on human pose estimation can be found in the literature, but they focus on a certain category; for example, model-based approaches or human motion analysis, etc. As far as we know, an overall review of this problem domain has yet to be provided. Furthermore, recent advancements based on deep learning have brought novel algorithms for this problem. In this paper, a comprehensive survey of human pose estimation from monocular images is carried out including milestone works and recent advancements. Based on one standard pipeline for the solution of computer vision problems, this survey splits the problem into several modules: feature extraction and description, human body models, and modeling methods. Problem modeling methods are approached based on two means of categorization in this survey. One way to categorize includes top-down and bottom-up methods, and another way includes generative and discriminative methods. Considering the fact that one direct application of human pose estimation is to provide initialization for automatic video surveillance, there are additional sections for motion-related methods in all modules: motion features, motion models, and motion-based methods. Finally, the paper also collects 26 publicly available data sets for validation and provides error measurement methods that are frequently used.

show abstract

Section: Methodologiesmentioning

confidence: 99%

Human Pose Estimation from Monocular Images: A Comprehensive Survey

Gong

Zhang

Gonzàlez

et al. 2016

Sensors

130

View full text Add to dashboard Cite

show abstract

“…The technique of the integration of global and local features has been successfully applied in image recognition, 21,25 and it improves the classification performance and enhances the robustness against interference, shielding and distortion.…”

Section: Pattern Recognition Based On Multiscale Sparse Representationmentioning

confidence: 99%

Sparse representation based on adaptive multiscale features for robust machinery fault diagnosis

Zhu

Wang

Zhao³

et al. 2014

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

In machinery fault diagnosis, it is fairly time consuming and expertise-demanded for manually selecting features, so it is profitable to automate this process for rapid and robust fault diagnosis. An automatic and adaptive feature extraction scheme via K-SVD algorithm was proposed in this paper, and without additional classifier, the fault detection was directly implemented by sparse representation. Higher animals apply the integration of global and local information to identify unknown objects for better recognition. Enlightened by this mechanism, the judgments by global and local frequency features were fused for better diagnosis by evidence theory. This fusion not only improved the successful rate, but also presented the reliability of diagnosis, which provided worthwhile recommendations and was essential to final decision. Verified in bearing fault diagnosis, the results demonstrated that the proposed scheme improved accuracy, robustness and efficiency, and this scheme had potential value for engineering application.

show abstract

“…Errave and time of sparse coding with different number of dictionary bases. (7). When α is fixed, this becomes a least-squares problem, and lots of solution are developed to solve it, like the K-SVD [17].…”

Section: B Fconvmentioning

confidence: 99%

Occluded Joints Recovery in 3D Human Pose Estimation based on Distance Matrix

Guo

Dai

2018

2018 24th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

Albeit the recent progress in single image 3D human pose estimation due to the convolutional neural network, it is still challenging to handle real scenarios such as highly occluded scenes. In this paper, we propose to address the problem of single image 3D human pose estimation with occluded measurements by exploiting the Euclidean distance matrix (EDM). Specifically, we present two approaches based on EDM, which could effectively handle occluded joints in 2D images. The first approach is based on 2D-to-2D distance matrix regression achieved by a simple CNN architecture. The second approach is based on sparse coding along with a learned over-complete dictionary. Experiments on the Human3.6M dataset show the excellent performance of these two approaches in recovering occluded observations and demonstrate the improvements in accuracy for 3D human pose estimation with occluded joints. 1

show abstract

Estimating Human Pose from Occluded Images

Cited by 21 publications

References 32 publications

Human Pose Estimation from Monocular Images: A Comprehensive Survey

Human Pose Estimation from Monocular Images: A Comprehensive Survey

Sparse representation based on adaptive multiscale features for robust machinery fault diagnosis

Occluded Joints Recovery in 3D Human Pose Estimation based on Distance Matrix

Contact Info

Product

Resources

About