Search

Bag-of-words (BoW) methods are a popular class of object recognition methods that use image features (e.g. SIFT) to form visual dictionaries and subsequent histogram vectors to represent object images in the recognition process. The accuracy of the BoW classifiers, however, is often limited by the presence of uninformative features extracted from the background or irrelevant image segments. Most existing solutions to prune out uninformative features rely on enforcing pairwise epipolar geometry via an expensive structure-from-motion (SfM) procedure. Such solutions are known to break down easily when the camera transformation is large or when the features are extracted from low-resolution low-quality images. In this paper, we propose a novel method to select informative object features using a more efficient algorithm called Sparse PCA. First, we show that using a large-scale multiple-view object database, informative features can be reliably identified from a high-dimensional visual dictionary by applying Sparse PCA on the histograms of each object category. Our experiment shows that the new algorithm improves recognition accuracy compared to the traditional BoW methods and SfM methods. Second, we present a new solution to Sparse PCA as a semidefinite programming problem using Augmented Lagrange Multiplier methods. The new solver outperforms the state of the art for estimating sparse principal vectors as a basis for a low-dimensional subspace model. The source code of our algorithms will be made public on our website., The original document contains color images.

Indoor localization is a challenging problem addressed extensively by both robotics and computer vision communities. Most existing approaches focus on using either cameras or laser scanners as the primary sensor for pose estimation. In laser scan matching based localization, finding scan point correspondences across scans is a challenging problem as individual scan points lack unique attributes. In camera based localization, one has to deal with images with little or no visual features as well as scale factor ambiguities to recover absolute distances. In this paper, we develop a multimodal approach for indoor localization by fusing a camera and a laser scanner in order to alleviate the drawbacks of each individual modality. Specifically, we use structure from motion to estimate the pose of a moving camera-laser rig which is subsequently used to compute scan correspondence estimates which are refined using a window based search method for scan point projections on the images. We have demonstrated our proposed system, consisting of a laser scanner and a camera, to result in a 0.4% loop closure error for a 60m loop around the interior corridor of a building.