Sharing Visual Features for Multiclass and Multiview Object Detection

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 29; no. 5; pp. 854 - 869
• Main Authors: Torralba, A., Murphy, K.P., Freeman, W.T.
• Format: Journal Article
• Language: English
• Published: Los Alamitos, CA IEEE 01.05.2007; IEEE Computer Society; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied sciences; Artificial Intelligence; Batteries; Battery; Boosting; Classifiers; Cluster Analysis; Complexity; Computation; Computational complexity; Computational efficiency; Computer science; control theory; systems; Costs; Detectors; Exact sciences and technology; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Information Storage and Retrieval - methods; interclass transfer; Layout; multiclass; Object detection; Pattern Recognition, Automated - methods; Pattern recognition. Digital image processing. Computational geometry; Reproducibility of Results; Runtime; Sensitivity and Specificity; sharing features; Studies; Subtraction Technique; Training; Training data; boosting; Decision making; interclass transfer; Secondary cell; Computational complexity; Clutter; Multiple image; Aggregate model; Multiple view; Image analysis; Proximity detector; sharing features; Supervised learning; Battery; Classification; Object detection; Localization; Pattern analysis; Time complexity; Artificial intelligence; Multiclass
• ISSN: 0162-8828; 2160-9292; 1939-3539
• DOI: 10.1109/TPAMI.2007.1055

We consider the problem of detecting a large number of different classes of objects in cluttered scenes. Traditional approaches require applying a battery of different classifiers to the image, at multiple locations and scales. This can be slow and can require a lot of training data since each classifier requires the computation of many different image features. In particular, for independently trained detectors, the (runtime) computational complexity and the (training-time) sample complexity scale linearly with the number of classes to be detected. We present a multitask learning procedure, based on boosted decision stumps, that reduces the computational and sample complexity by finding common features that can be shared across the classes (and/or views). The detectors for each class are trained jointly, rather than independently. For a given performance level, the total number of features required and, therefore, the runtime cost of the classifier, is observed to scale approximately logarithmically with the number of classes. The features selected by joint training are generic edge-like features, whereas the features chosen by training each class separately tend to be more object-specific. The generic features generalize better and considerably reduce the computational cost of multiclass object detection