How important are 'Deformable Parts' in the Deformable Parts Model?

The Deformable Parts Model (DPM) has recently emerged as a very useful and popular tool for tackling the intra-category diversity problem in object detection. In this paper, we summarize the key insights from our empirical analysis of the important elements constituting this detector. More specifically, we study the relationship between the role of deformable parts and the mixture model components within this de- tector, and understand their relative importance. First, we find that by increasing the number of components, and switching the initialization step from their aspect-ratio, left-right flipping heuristics to appearance- based clustering, considerable improvement in performance is obtained. But more intriguingly, we observed that with these new components, the part deformations can now be turned off, yet obtaining results that are almost on par with the original DPM detector.


Appearance-based subcategory detection


"How important are 'Deformable Parts' in the Deformable Parts Model?"
Santosh K. Divvala, Alexei A. Efros, Martial Hebert
Parts and Attributes Workshop, ECCV 2012
[Paper] [Presentation] [Bibtex]
[Subcategories discovered for the 20 VOC categories]

"Context and Subcategories for Sliding Window Object Recognition"
Santosh K. Divvala
Doctoral Dissertation, Robotics Institute, Carnegie Mellon University, August 2012
[Announcement] [Thesis Document]