0 presente trabalho propõe metodos para localizar e rastrear indivfduos combinando evidencia oriunda de multiplas cameras, atraves da restrigao homografica induzida pelo piano do solo. Os procedimentos propostos utilizam um subtrator de fundo para definir quais pixels pertencem aos objetos de interesse. Esses pixeis sao empregados como evidencia da localizagao de pessoas no piano de referencia. Os algoritmos propostos computam a quantidade de suporte, que corresponde a 'massa' observada acima de cada pixel. Pixeis que correspondem as localizag6es no solo onde se encontram os indivfduos irao apresentar maior suporte. Esse suporte e normalizado para compensar efeitos de perspectiva e acumulado no piano de referenda para todas as cameras observadas. A detecgao de pessoas no piano do solo torna-se o problema de busca por regi6es de maximos locais no acumulador. Falsos-positivos sao filtrados atraves de uma avaliagao de consisten- cia entre os candidates encontrados. Os candidates remanescentes sao rastreados atraves de Filtros de Kalman e um modelo de aparencia multicamera. Resultados experimentais a partir de dados provenientes

Title in English

not available

Abstract in English

This paper proposes a method to locate and track people by combining evidence from multiple cameras using the homography constraint. The proposed method use foreground pixels from simple background subtraction to compute evidence of the location of people on a reference ground plane. The algorithm computes the amount of support that basically corresponds to the 'foreground mass' above each pixel. Therefore, pixels that correspond to ground points have more support. The support is normalized to compensate for perspective effects and accumulated on the reference plane for all camera views. The detection of people on the reference plane becomes a search for regions of local maxima in the accumulator. Many false positives are filtered by checking the visibility consistency of the detected candidates against all camera views. The remaining candidates are tracked using Kalman filters and appearance models. Experimental results using challenging data from PETS'06 and PETS'09 show good performance of the method in the presence of severe occlusion.

 

WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.

SantosThiagoTeixeira.pdf (8.39 Mbytes)

Publishing Date

2023-07-27

 

Derived works

WARNING: Learn what derived works are clicking here.