We introduce the Generative Template of Features (GTF), a parts-based model for visual object category detection. The GTF consists of a number of parts, and for each part there is a corresponding spatial location distribution and a distribution over ‘visual words’ (clusters of invariant features). The performance of the GTF is evaluated for object localisation, and it is shown that such a relatively simple model can give state-of-the-art performance. We also demonstrate how a Hough-transform-like method for object localisation can be derived from the GTF model.
- Object localisation
- Visual words
- Generative Template of Features
- Object recognition