A tensor based on optical flow for global description of motion in videos

Virginia Fernandes Mota (1) , Eder de Almeida Perez (1) , Marcelo Bernardes Vieira (1) , Silva Maciel Luiz Maurílio Da (1) , Frédéric Precioso (2) , Philippe-Henri Gosselin (3)

1 DCC / ICE - UFJF - Departamento de Ciência da Computação [Juiz de Fora]
2 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe KEIA
3 MIDI - Multimedia Indexation and Data Integration

Motion is one of the main characteristics that describe the semantic information of videos. In this work, a global video descriptor based on orientation tensors is proposed. This descriptor is obtained by combining polynomial coefficients calculated for each image in a video. The coefficients are found through the projection of the optical flow on Legendre polyno- mials, reducing the dimension of per frame motion estimations. The sequence of coefficients are then combined using orientation tensors. The global tensor descriptor created is evaluated by a classification of the KTH video database with a SVM classifier.

Mots clés: Global descriptor; Orientation tensor; SVM; Optical Flow; Motion modeling

Domaines: Multimédia [cs.MM]; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Virginia Fernandes Mota, Eder de Almeida Perez, Marcelo Bernardes Vieira, Silva Maciel Luiz Maurílio Da, Frédéric Precioso, et al.. A tensor based on optical flow for global description of motion in videos. SIBGRAPI 2012, Aug 2012, Ouro Preto, Brazil. pp.4. ⟨hal-00753159⟩