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. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1007/s11263-013-0670-8 | International Journal of Computer Vision |
Keywords | Field | DocType |
Branch&rank,Object detection,Non-linear kernel classifier,Sub-linear detection | Structured support vector machine,Object detection,Ranking SVM,Pattern recognition,Ranking,Computer science,Support vector machine,Artificial intelligence,Classifier (linguistics),Contextual image classification,Machine learning,Cognitive neuroscience of visual object recognition | Journal |
Volume | Issue | ISSN |
106 | 3 | 0920-5691 |
Citations | PageRank | References |
2 | 0.39 | 47 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Alain Lehmann | 1 | 78 | 4.42 |
Peter Gehler | 2 | 1363 | 61.64 |
Luc Van Gool | 3 | 27566 | 1819.51 |