Branch&Rank for Efficient Object Detection

  • Authors:

  • Alain D. Lehmann;Peter V. Gehler;Luc Van Gool

  • Affiliations:

  • Computer Vision Laboratory, ETH Zurich, Zurich, Switzerland;MPI for Intelligent Systems, Tübingen, Germany;Computer Vision Laboratory, ETH Zurich, Zurich, Switzerland and ESAT-PSI/IBBT, KU Leuven, Leuven, Belgium

  • Venue:
  • International Journal of Computer Vision
  • Year:
  • 2014

Quantified Score

Hi-index 0.00

Visualization

Abstract

Ranking hypothesis sets is a powerful concept for efficient object detection. In this work, we propose a branch&rank scheme that detects objects with often less than 100 ranking operations. This efficiency enables the use of strong and also costly classifiers like non-linear SVMs with RBF- $$\chi ^2$$ 驴 2 kernels. We thereby relieve an inherent limitation of branch&bound methods as bounds are often not tight enough to be effective in practice. Our approach features three key components: a ranking function that operates on sets of hypotheses and a grouping of these into different tasks. Detection efficiency results from adaptively sub-dividing the object search space into decreasingly smaller sets. This is inherited from branch&bound, while the ranking function supersedes a tight bound which is often unavailable (except for rather limited function classes). The grouping makes the system effective: it separates image classification from object recognition, yet combines them in a single formulation, phrased as a structured SVM problem. A novel aspect of branch&rank is that a better ranking function is expected to decrease the number of classifier calls during detection. We use the VOC'07 dataset to demonstrate the algorithmic properties of branch&rank.