Semi-Supervised Learning for Location Recognition from Wearable Video

Vladislavs Dovgalecs (1) , Rémi Megret (1) , Hazem Wannous (1) , Yannick Berthoumieu (1)

1 IMS - Laboratoire de l'intégration, du matériau au système

This paper tackles the problem of image-based in- door location recognition. The context of the present work is activity monitoring using a wearable video cam- era data. Because application constraints necessitate weak supervision, a semi-supervised approach has been adopted which leverages the large amount of unlabeled images. The proposed method is based on the Bag of Features approach for image description followed by spectral dimensionality reduction in a transductive setup. Additional information from geometrical veri- fication constraints are also considered which allowed to reach higher performance levels. The considered al- gorithms are compared experimentally on the data ac- quired in the wearable camera setup.

Mots clés: location recognition; wearable video; semi-supervised learning; dimensionnality reduction; bag-of-features; interest point matching

Domaines: Traitement du signal et de l'image [eess.SP]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Traitement du signal et de l'image [eess.SP]

Vladislavs Dovgalecs, Rémi Megret, Hazem Wannous, Yannick Berthoumieu. Semi-Supervised Learning for Location Recognition from Wearable Video. International Workshop on Content-Based Multimedia Indexing (CBMI), Jun 2010, Grenoble, France. ⟨10.1109/CBMI.2010.5529903⟩. ⟨hal-00547964⟩