Learn More
We present a family of sliding window object detectors that combine a rich visual feature set with methodological advances from [2, 4, 5], giving state-of-the-art performance on several important datasets. Using the Latent SVM learning framework [4], our basic root detectors outperform the single component part-based ones of Felzenszwalb et. al on 9 of 10(More)
This paper proposes a new image representation for texture categorization and facial analysis, relying on the use of higher-order local differential statistics as features. In contrast with models based on the global structure of textures and faces, it has been shown recently that small local pixel pattern distributions can be highly discrimina-tive.(More)
  • 1