Hispana. Acceso en línea al Patrimonio Cultural Digital Español

Está en:  › Dades de registre

3D human pose estimation from depth maps using a deep combination of poses

Identificadores del recurso

Marín-Jiménez MJ, Romero-Ramirez FJ, Muñoz-Salinas R, Medina-Carnicer R. 3D human pose estimation from depth maps using a deep combination of poses. Journal Of Visual Communication And Image Representation [Internet]. 1 de agosto de 2018;55:627-39.

1047-3203

http://hdl.handle.net/10668/18831

10.1016/j.jvcir.2018.07.010

1095-9076

http://arxiv.org/pdf/1807.05389

445318100053

Procedència

(RISalud-ANDALUCÍA)

Fitxa

Títol:: 3D human pose estimation from depth maps using a deep combination of poses
Tema:: 3D human pose; Body limbs; Depth maps; ConvNets; Deep learning; Human body; Neural networks, computer; Image processing, computer-assisted; Aprendizaje profundo; Cuerpo humano; Procesamiento de imagen asistido por computador; Redes neurales de la computación
Descripció:: Many real-world applications require the estimation of human body joints for higher-level tasks as, for example, human behaviour understanding. In recent years, depth sensors have become a popular approach to obtain three-dimensional information. The depth maps generated by these sensors provide information that can be employed to disambiguate the poses observed in two-dimensional images. This work addresses the problem of 3D human pose estimation from depth maps employing a Deep Learning approach. We propose a model, named Deep Depth Pose (DDP), which receives a depth map containing a person and a set of predefined 3D prototype poses and returns the 3D position of the body joints of the person. In particular, DDP is defined as a ConvNet that computes the specific weights needed to linearly combine the prototypes for the given input. We have thoroughly evaluated DDP on the challenging 'ITOP' and 'UBC3V' datasets, which respectively depict realistic and synthetic samples, defining a new state-of-the-art on them.; This project has been funded under projects TIN2016-75279-P and IFI16/00033 (ISCIII) of Spain Ministry of Economy, Industry and Competitiveness, and FEDER. Thanks to NVidia for donating the GPU Titan Xp used for the experiments presented in this work. We also thank Shafaei and Little for providing their error and precision results used in our comparative plots.; Si
Idioma:
Relació:: https://www.sciencedirect.com/science/article/abs/pii/S1047320318301718; TIN2016-75279-P; IFI16/00033
Autor/Productor:: Marin-Jimenez, Manuel J.; Romero-Ramirez, Francisco J.; Munoz-Salinas, Rafael; Medina-Carnicer, Rafael
Editor:: Academic Press
Otros colaboradores/productores:: [Marin-Jimenez, Manuel J.] Univ Cordoba, Dept Informat & Anal Numer, Campus Rabanales, E-14071 Cordoba, Spain; [Romero-Ramirez, Francisco J.] Univ Cordoba, Dept Informat & Anal Numer, Campus Rabanales, E-14071 Cordoba, Spain; [Munoz-Salinas, Rafael] Univ Cordoba, Dept Informat & Anal Numer, Campus Rabanales, E-14071 Cordoba, Spain; [Medina-Carnicer, Rafael] Univ Cordoba, Dept Informat & Anal Numer, Campus Rabanales, E-14071 Cordoba, Spain; [Marin-Jimenez, Manuel J.] Inst Maimonides Invest Biomed IMIBIC, Ave Menendez Pida1 S-N, Cordoba 14004, Spain; [Munoz-Salinas, Rafael] Inst Maimonides Invest Biomed IMIBIC, Ave Menendez Pida1 S-N, Cordoba 14004, Spain; [Medina-Carnicer, Rafael] Inst Maimonides Invest Biomed IMIBIC, Ave Menendez Pida1 S-N, Cordoba 14004, Spain; (ISCIII) of Spain Ministry of Economy, Industry and Competitiveness; FEDER; Spain Ministry of Economy, Industry and Competitiveness
Drets:: open access
Data:: 2023-02-12T02:20:59Z; 2018-07-17
Tipo de recurso:: Research article; SMUR
Format:: application/pdf

oai_dc

Descarregar XML

<?xml version="1.0" encoding="UTF-8" ?>

<oai_dc:dc schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
1. <dc:title>3D human pose estimation from depth maps using a deep combination of poses</dc:title>
2. <dc:creator>Marin-Jimenez, Manuel J.</dc:creator>
3. <dc:creator>Romero-Ramirez, Francisco J.</dc:creator>
4. <dc:creator>Munoz-Salinas, Rafael</dc:creator>
5. <dc:creator>Medina-Carnicer, Rafael</dc:creator>
6. <dc:contributor>[Marin-Jimenez, Manuel J.] Univ Cordoba, Dept Informat & Anal Numer, Campus Rabanales, E-14071 Cordoba, Spain</dc:contributor>
7. <dc:contributor>[Romero-Ramirez, Francisco J.] Univ Cordoba, Dept Informat & Anal Numer, Campus Rabanales, E-14071 Cordoba, Spain</dc:contributor>
8. <dc:contributor>[Munoz-Salinas, Rafael] Univ Cordoba, Dept Informat & Anal Numer, Campus Rabanales, E-14071 Cordoba, Spain</dc:contributor>
9. <dc:contributor>[Medina-Carnicer, Rafael] Univ Cordoba, Dept Informat & Anal Numer, Campus Rabanales, E-14071 Cordoba, Spain</dc:contributor>
10. <dc:contributor>[Marin-Jimenez, Manuel J.] Inst Maimonides Invest Biomed IMIBIC, Ave Menendez Pida1 S-N, Cordoba 14004, Spain</dc:contributor>
11. <dc:contributor>[Munoz-Salinas, Rafael] Inst Maimonides Invest Biomed IMIBIC, Ave Menendez Pida1 S-N, Cordoba 14004, Spain</dc:contributor>
12. <dc:contributor>[Medina-Carnicer, Rafael] Inst Maimonides Invest Biomed IMIBIC, Ave Menendez Pida1 S-N, Cordoba 14004, Spain</dc:contributor>
13. <dc:contributor>(ISCIII) of Spain Ministry of Economy, Industry and Competitiveness</dc:contributor>
14. <dc:contributor>FEDER</dc:contributor>
15. <dc:contributor>Spain Ministry of Economy, Industry and Competitiveness</dc:contributor>
16. <dc:contributor>FEDER</dc:contributor>
17. <dc:subject>3D human pose</dc:subject>
18. <dc:subject>Body limbs</dc:subject>
19. <dc:subject>Depth maps</dc:subject>
20. <dc:subject>ConvNets</dc:subject>
21. <dc:subject>Deep learning</dc:subject>
22. <dc:subject>Human body</dc:subject>
23. <dc:subject>Neural networks, computer</dc:subject>
24. <dc:subject>Image processing, computer-assisted</dc:subject>
25. <dc:subject>Aprendizaje profundo</dc:subject>
26. <dc:subject>Cuerpo humano</dc:subject>
27. <dc:subject>Procesamiento de imagen asistido por computador</dc:subject>
28. <dc:subject>Redes neurales de la computación</dc:subject>
29. <dc:description>Many real-world applications require the estimation of human body joints for higher-level tasks as, for example, human behaviour understanding. In recent years, depth sensors have become a popular approach to obtain three-dimensional information. The depth maps generated by these sensors provide information that can be employed to disambiguate the poses observed in two-dimensional images. This work addresses the problem of 3D human pose estimation from depth maps employing a Deep Learning approach. We propose a model, named Deep Depth Pose (DDP), which receives a depth map containing a person and a set of predefined 3D prototype poses and returns the 3D position of the body joints of the person. In particular, DDP is defined as a ConvNet that computes the specific weights needed to linearly combine the prototypes for the given input. We have thoroughly evaluated DDP on the challenging 'ITOP' and 'UBC3V' datasets, which respectively depict realistic and synthetic samples, defining a new state-of-the-art on them.</dc:description>
30. <dc:description>This project has been funded under projects TIN2016-75279-P and IFI16/00033 (ISCIII) of Spain Ministry of Economy, Industry and Competitiveness, and FEDER. Thanks to NVidia for donating the GPU Titan Xp used for the experiments presented in this work. We also thank Shafaei and Little for providing their error and precision results used in our comparative plots.</dc:description>
31. <dc:description>Si</dc:description>
32. <dc:date>2023-02-12T02:20:59Z</dc:date>
33. <dc:date>2023-02-12T02:20:59Z</dc:date>
34. <dc:date>2018-07-17</dc:date>
35. <dc:type>Research article</dc:type>
36. <dc:type>SMUR</dc:type>
37. <dc:identifier>Marín-Jiménez MJ, Romero-Ramirez FJ, Muñoz-Salinas R, Medina-Carnicer R. 3D human pose estimation from depth maps using a deep combination of poses. Journal Of Visual Communication And Image Representation [Internet]. 1 de agosto de 2018;55:627-39.</dc:identifier>
38. <dc:identifier>1047-3203</dc:identifier>
39. <dc:identifier>http://hdl.handle.net/10668/18831</dc:identifier>
40. <dc:identifier>10.1016/j.jvcir.2018.07.010</dc:identifier>
41. <dc:identifier>1095-9076</dc:identifier>
42. <dc:identifier>http://arxiv.org/pdf/1807.05389</dc:identifier>
43. <dc:identifier>445318100053</dc:identifier>
44. <dc:language>en</dc:language>
45. <dc:relation>https://www.sciencedirect.com/science/article/abs/pii/S1047320318301718</dc:relation>
46. <dc:relation>TIN2016-75279-P</dc:relation>
47. <dc:relation>IFI16/00033</dc:relation>
48. <dc:rights>open access</dc:rights>
49. <dc:format>application/pdf</dc:format>
50. <dc:publisher>Academic Press</dc:publisher>
</oai_dc:dc>