The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Propomos uma abordagem nova e eficiente para rastrear partes articuladas do corpo humano em 2D. Na nossa abordagem, o corpo humano é modelado por um modelo gráfico onde cada parte é representada por um nó e a relação entre um par de partes adjacentes é indicada por uma aresta no gráfico. Várias abordagens foram propostas para resolver tais problemas, mas a eficiência ainda é um problema vital. Apresentamos uma nova abordagem baseada em Quick Shift Belief Propagation (QSBP) que se beneficia do Quick Shift, um método simples e eficiente de busca de modo, em um modelo de propagação de crenças baseado em partes. O aspecto único deste modelo é a sua capacidade de descobrir eficientemente os modos da distribuição de probabilidade marginal subjacente, preservando ao mesmo tempo a precisão. Isso dá ao QSBP uma vantagem significativa sobre abordagens como Propagação de Crenças (BP) e Propagação de Crenças por Mudança Média (MSBP). Além disso, demonstramos o uso do QSBP com um modelo baseado em ação; isso proporciona vantagens adicionais de lidar com a autooclusão e reduzir ainda mais o espaço de busca. Apresentamos análises qualitativas e quantitativas da abordagem proposta com resultados encorajadores.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Kittiya KHONGKRAPHAN, Pakorn KAEWTRAKULPONG, "Efficient Human Body Tracking by Quick Shift Belief Propagation" in IEICE TRANSACTIONS on Information,
vol. E94-D, no. 4, pp. 905-912, April 2011, doi: 10.1587/transinf.E94.D.905.
Abstract: We propose a novel and efficient approach for tracking 2D articulated human body parts. In our approach, the human body is modeled by a graphical model where each part is represented by a node and the relationship between a pair of adjacent parts is indicated by an edge in the graph. Various approaches have been proposed to solve such problems, but efficiency is still a vital problem. We present a new Quick Shift Belief Propagation (QSBP) based approach which benefits from Quick Shift, a simple and efficient mode seeking method, in a part based belief propagation model. The unique aspect of this model is its ability to efficiently discover modes of the underlying marginal probability distribution while preserving the accuracy. This gives QSBP a significant advantage over approaches like Belief Propagation (BP) and Mean Shift Belief Propagation (MSBP). Moreover, we demonstrate the use of QSBP with an action based model; this provides additional advantages of handling self-occlusion and further reducing the search space. We present qualitative and quantitative analysis of the proposed approach with encouraging results.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E94.D.905/_p
Copiar
@ARTICLE{e94-d_4_905,
author={Kittiya KHONGKRAPHAN, Pakorn KAEWTRAKULPONG, },
journal={IEICE TRANSACTIONS on Information},
title={Efficient Human Body Tracking by Quick Shift Belief Propagation},
year={2011},
volume={E94-D},
number={4},
pages={905-912},
abstract={We propose a novel and efficient approach for tracking 2D articulated human body parts. In our approach, the human body is modeled by a graphical model where each part is represented by a node and the relationship between a pair of adjacent parts is indicated by an edge in the graph. Various approaches have been proposed to solve such problems, but efficiency is still a vital problem. We present a new Quick Shift Belief Propagation (QSBP) based approach which benefits from Quick Shift, a simple and efficient mode seeking method, in a part based belief propagation model. The unique aspect of this model is its ability to efficiently discover modes of the underlying marginal probability distribution while preserving the accuracy. This gives QSBP a significant advantage over approaches like Belief Propagation (BP) and Mean Shift Belief Propagation (MSBP). Moreover, we demonstrate the use of QSBP with an action based model; this provides additional advantages of handling self-occlusion and further reducing the search space. We present qualitative and quantitative analysis of the proposed approach with encouraging results.},
keywords={},
doi={10.1587/transinf.E94.D.905},
ISSN={1745-1361},
month={April},}
Copiar
TY - JOUR
TI - Efficient Human Body Tracking by Quick Shift Belief Propagation
T2 - IEICE TRANSACTIONS on Information
SP - 905
EP - 912
AU - Kittiya KHONGKRAPHAN
AU - Pakorn KAEWTRAKULPONG
PY - 2011
DO - 10.1587/transinf.E94.D.905
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E94-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2011
AB - We propose a novel and efficient approach for tracking 2D articulated human body parts. In our approach, the human body is modeled by a graphical model where each part is represented by a node and the relationship between a pair of adjacent parts is indicated by an edge in the graph. Various approaches have been proposed to solve such problems, but efficiency is still a vital problem. We present a new Quick Shift Belief Propagation (QSBP) based approach which benefits from Quick Shift, a simple and efficient mode seeking method, in a part based belief propagation model. The unique aspect of this model is its ability to efficiently discover modes of the underlying marginal probability distribution while preserving the accuracy. This gives QSBP a significant advantage over approaches like Belief Propagation (BP) and Mean Shift Belief Propagation (MSBP). Moreover, we demonstrate the use of QSBP with an action based model; this provides additional advantages of handling self-occlusion and further reducing the search space. We present qualitative and quantitative analysis of the proposed approach with encouraging results.
ER -