Viewpoint-Independent Object Class Detection using 3D Feature Maps

Jörg Liebelt (1) , Cordelia Schmid (2) , Klaus Schertler (1)

1 EADS IW - EADS Innovation Works [Munich]
2 LEAR - Learning and recognition in vision

This paper presents a 3D approach to multi-view object class detection. Most existing approaches recognize object classes for a particular viewpoint or combine classifiers for a few discrete views. We propose instead to build 3D representations of object classes which allow to handle viewpoint changes and intra-class variability. Our approach extracts a set of pose and class discriminant features from synthetic 3D object models using a filtering procedure, evaluates their suitability for matching to real image data and represents them by their appearance and 3D position. We term these representations 3D Feature Maps. For recognizing an object class in an image we match the synthetic descriptors to the real ones in a 3D voting scheme. Geometric coherence is reinforced by means of a robust pose estimation which yields a 3D bounding box in addition to the 2D localization. The precision of the 3D pose estimation is evaluated on a set of images of a calibrated scene. The 2D localization is evaluated on the PASCAL 2006 dataset for motorbikes and cars, showing that its performance can compete with state-of-the-art 2D object detectors.

Mots clés: feature extraction; filtering theory; image classification; image representation; object detection; object recognition; pose estimation

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Jörg Liebelt, Cordelia Schmid, Klaus Schertler. Viewpoint-Independent Object Class Detection using 3D Feature Maps. CVPR 2008 - IEEE Conference on Computer Vision & Pattern Recognition, Jun 2008, Anchorage, United States. pp.1-8, ⟨10.1109/CVPR.2008.4587614⟩. ⟨inria-00548657⟩