Timely Object Recognition
A system for deploying object detectors in sequence, guided by a policy that takes into account the current "belief state" of the system. The policy is learned using reinforcement learning techniques.
Configured to use pre-output DPM  detections and pre-output GIST features on the PASCAL VOC dataset. Relies on the FastInf library  for one of the inference modes (code included).
This code accompanies my NIPS 2012 publication on Timely Object Recognition.
For more recent code, see the project page.
-  P. F. Felzenszwalb, R. B. Girshick, and D. McAllester, "Cascade object detection with deformable part models," in CVPR, 2010.
-  A. Jaimovich and I. Mcgraw, "FastInf : An Efficient Approximate Inference Library," Journal of Machine Learning Research, vol. 11, pp. 1733–1736, 2010.
-  S. Karayev, T. Baumgartner, M. Fritz, and T. Darrell, "Timely Object Recognition," in NIPS, 2012.