Português English
Contato

Dissertação de Leonardo Perdomo


Detalhes do Evento


Aluno: Leonardo Perdomo
Orientador: Edson Prestes e Silva Junior

Título: c-M2DP: A Fast Point Cloud Descriptor with Color Information to Perform Loop Closure Detection
Linha de Pesquisa: Planejamento, Sistemas Multiagentes e Robótica

Data: 11/11/2019
Hora: 13:30
Local: Prédio 43412 – Sala 215 – Instituto de Informática UFRGS

Banca Examinadora:
Prof. Dr. Claudio Rosito Jung (UFRGS)
Prof. Dr. Edison Pignaton de Freitas (UFRGS)
Prof. Dr. Alexandre de Morais Amory (PUCRS)

Presidente da Banca: Prof. Dr. Edson Prestes e Silva Junior

Abstract: We present c-M2DP, a fast global point cloud descriptor that combines color and shape information, and perform loop closure detection using it. Our approach extends the M2DP descriptor by incorporating color information. Along with M2DP shape signatures, we compute color signatures from multiple 2D projections of a point cloud. Then, a compact descriptor is computed using SVD to reduce its dimensionality. We performed experiments on four public available dataset sequences, with point clouds generated using either camera-LIDAR fusion or stereo depth estimation. Our results show an overall accuracy improvement over M2DP while maintaining efficiency, and are competitive in comparison with another color and shape descriptor.

Keywords: Point Cloud Descriptor, Loop Closure Detection, SLAM.