A non-myopic approach to visual search

We show how a greedy approach to visual search - i.e., directly moving to the most likely location of the target - can be suboptimal, if the target object is hard to detect. Instead it is more efficient and leads to higher detection accuracy to first look for other related objects, that are easier to detect. These provide contextual priors for the target that make it easier to find. We demonstrate this in simulation using POMDP models, focussing on two special cases: where the target object is contained within the related object, and where the target object is spatially adjacent to the related object.

[1]  Yishay Mansour,et al.  A Sparse Sampling Algorithm for Near-Optimal Planning in Large Markov Decision Processes , 1999, Machine Learning.

[2]  Nikos A. Vlassis,et al.  Perseus: Randomized Point-based Value Iteration for POMDPs , 2005, J. Artif. Intell. Res..

[3]  Daniel P. Huttenlocher,et al.  Pictorial Structures for Object Recognition , 2004, International Journal of Computer Vision.

[4]  Christopher M. Brown,et al.  Control of selective perception using bayes nets and decision theory , 1994, International Journal of Computer Vision.

[5]  Frank P. Ferrie,et al.  Active Object Recognition: Looking for Differences , 2001, International Journal of Computer Vision.

[6]  Andrew G. Barto,et al.  Learning to Act Using Real-Time Dynamic Programming , 1995, Artif. Intell..

[7]  Antonio Torralba,et al.  Using the Forest to See the Trees: A Graphical Model Relating Features, Objects, and Scenes , 2003, NIPS.

[8]  Steffen L. Lauritzen,et al.  Representing and Solving Decision Problems with Limited Information , 2001, Manag. Sci..

[9]  Leslie Pack Kaelbling,et al.  Planning and Acting in Partially Observable Stochastic Domains , 1998, Artif. Intell..

[10]  Lambert E. Wixson,et al.  Using intermediate objects to improve the efficiency of visual search , 1994, International Journal of Computer Vision.

[11]  Tal Arbel,et al.  Efficient Discriminant Viewpoint Selection for Active Bayesian Recognition , 2006, International Journal of Computer Vision.

[12]  Pietro Perona,et al.  A sparse object category model for efficient learning and exhaustive recognition , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[13]  Hector Geffner,et al.  Solving Large POMDPs using Real Time Dynamic Programming , 1998 .

[14]  James H. Elder,et al.  Pre-Attentive and Attentive Detection of Humans in Wide-Field Scenes , 2007, International Journal of Computer Vision.

[15]  Finn V. Jensen,et al.  Bayesian Networks and Decision Graphs , 2001, Statistics for Engineering and Information Science.

[16]  Peter Norvig,et al.  Artificial Intelligence: A Modern Approach , 1995 .

[17]  Kevin Murphy,et al.  Bayes net toolbox for Matlab , 1999 .

[18]  Ronald A. Rensink.  The Dynamic Representation of Scenes , 2000 .

[19]  Lucas Paletta,et al.  Active object recognition by view integration and reinforcement learning , 2000, Robotics Auton. Syst..

[20]  Daniel D. Fu,et al.  Vision and navigation in man-made environments: looking for syrup in all the right places , 1994 .

[21]  Eric Horvitz,et al.  Selective perception policies for guiding sensing and computation in multimodal systems: a comparative analysis , 2003, ICMI '03.

[22]  Sridhar Mahadevan,et al.  A reinforcement learning model of selective visual attention , 2001, AGENTS '01.

[23]  Fourth Canadian Conference on Computer and Robot Vision (CRV 2007), 28-30 May 2007, Montreal, Quebec, Canada , 2007, CRV.

[24]  Leslie Pack Kaelbling,et al.  Representing hierarchical POMDPs as DBNs for multi-scale robot localization , 2004, IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004.

[25]  Michael Fink,et al.  The Full Images for Natural Knowledge Caltech Office DB , 2003 .

[26]  Joelle Pineau,et al.  Point-based value iteration: An anytime algorithm for POMDPs , 2003, IJCAI.

[27]  Jesse Hoey,et al.  Solving POMDPs with Continuous or Large Discrete Observation Spaces , 2005, IJCAI.