Showing papers by "David G. Lowe published in 2011"

REIN - A fast, robust, scalable REcognition INfrastructure

Abstract: A robust robot perception system intended to enable object manipulation needs to be able to accurately identify objects and their pose at high speeds. Since objects vary considerably in surface properties, rigidity and articulation, no single detector or object estimation method has been shown to provide reliable detection across object types to date. This indicates the need for an architecture that is able to quickly swap detectors, pose estimators, and filters, or to run them in parallel or serial and combine their results, preferably without any code modifications at all. In this paper, we present our implementation of such an infrastructure, ReIn (REcognition INfrastructure), to answer these needs. ReIn is able to combine a multitude of 2D/3D object recognition and pose estimation techniques in parallel as dynamically loadable plugins. It also provides an extremely efficient data passing architecture, and offers the possibility to change the parameters and initial settings of these techniques during their execution. In the course of this work we introduce two new classifiers designed for robot perception needs: BiGGPy (Binarized Gradient Grid Pyramids) for scalable 2D classification and VFH (Viewpoint Feature Histograms) for 3D classification and pose. We then show how these two classifiers can be easily combined using ReIn to solve object recognition and pose identification problems.

Local Naive Bayes Nearest Neighbor for Image Classification

Abstract: We present Local Naive Bayes Nearest Neighbor, an improvement to the NBNN image classification algorithm that increases classification accuracy and improves its ability to scale to large numbers of object classes. The key observation is that only the classes represented in the local neighborhood of a descriptor contribute significantly and reliably to their posterior probability estimates. Instead of maintaining a separate search structure for each class, we merge all of the reference data together into one search structure, allowing quick identification of a descriptor's local neighborhood. We show an increase in classification accuracy when we ignore adjustments to the more distant classes and show that the run time grows with the log of the number of classes rather than linearly in the number of classes as did the original. This gives a 100 times speed-up over the original method on the Caltech 256 dataset. We also provide the first head-to-head comparison of NBNN against spatial pyramid methods using a common set of input features. We show that local NBNN outperforms all previous NBNN based methods and the original spatial pyramid model. However, we find that local NBNN, while competitive with, does not beat state-of-the-art spatial pyramid methods that use local soft assignment and max-pooling.

An adaptive interface for active localization

Abstract: Thanks to large-scale image repositories, vast amounts of data for object recognition are now easily available. However, acquiring training labels for arbitrary objects still requires tedious and expensive human effort. This is particularly true for localization, where humans must not only provide labels, but also training windows in an image. We present an approach for reducing the number of labelled training instances required to train an object classifier and for assisting the user in specifying optimal object location windows. As part of this process, the algorithm performs localization to find bounding windows for training examples that are best aligned with the current classification function, which optimizes learning and reduces human effort. To test this approach, we introduce an active learning extension to a latent SVM learning algorithm. Our user interface for training object detectors employs real-time interaction with a human user. Our active learning system provides a mean performance improvement of 4.5% in the average precision over a state of the art detector on the PASCAL Visual Object Classes Challenge 2007 with an average of just 40 minutes of human labelling effort per class.