Contextual visual localization: cascaded submap classification, optimized saliency detection, and fast view matching

In this paper, we present a novel coarse-to-fine visual localization approach: contextual visual localization. This approach relies on three elements: (i) a minimal-complexity classifier for performing fast coarse localization (submap classification); (ii) an optimized saliency detector which exploits the visual statistics of the submap; and (iii) a fast view-matching algorithm which filters initial matchings with a structural criterion. The latter algorithm yields fine localization. Our experiments show that these elements have been successfully integrated for solving the global localization problem. Context, that is, the awareness of being in a particular submap, is defined by a supervised classifier tuned for a minimal set of features. Visual context is exploited both for tuning (optimizing) the saliency detection process, and to select potential matching views in the visual database, close enough to the query view.

Expression and cellular localization of the voltage-gated calcium channel α2δ3 in the rodent retina

High-voltage-activated calcium channels are hetero-oligomeric protein complexes that mediate multiple cellular processes, including the influx of extracellular Ca2+, neurotransmitter release, gene transcription, and synaptic plasticity. These channels consist of a primary α1 pore-forming subunit, which is associated with an extracellular α2δ subunit and an intracellular β auxiliary subunit, which alter the gating properties and trafficking of the calcium channel. The cellular localization of the α2δ3 subunit in the mouse and rat retina is unknown. In this study using RT-PCR, a single band at ∼305 bp corresponding to the predicted size of the α2δ3 subunit fragment was found in mouse and rat retina and brain homogenates. Western blotting of rodent retina and brain homogenates showed a single 123-kDa band. Immunohistochemistry with an affinity-purified antibody to the α2δ3 subunit revealed immunoreactive cell bodies in the ganglion cell layer and inner nuclear layer and immunoreactive processes in the inner plexiform layer and the outer plexiform layer. α2δ3 immunoreactivity was localized to multiple cell types, including ganglion, amacrine, and bipolar cells and photoreceptors, but not horizontal cells. The expression of the α2δ3 calcium channel subunit to multiple cell types suggests that this subunit participates widely in Ca-channel-mediated signaling in the retina.

Semantic localization in the PCL library

The semantic localization problem in robotics consists in determining the place where a robot is located by means of semantic categories. The problem is usually addressed as a supervised classification process, where input data correspond to robot perceptions while classes to semantic categories, like kitchen or corridor. In this paper we propose a framework, implemented in the PCL library, which provides a set of valuable tools to easily develop and evaluate semantic localization systems. The implementation includes the generation of 3D global descriptors following a Bag-of-Words approach. This allows the generation of fixed-dimensionality descriptors from any type of keypoint detector and feature extractor combinations. The framework has been designed, structured and implemented to be easily extended with different keypoint detectors, feature extractors as well as classification models. The proposed framework has also been used to evaluate the performance of a set of already implemented descriptors, when used as input for a specific semantic localization system. The obtained results are discussed paying special attention to the internal parameters of the BoW descriptor generation process. Moreover, we also review the combination of some keypoint detectors with different 3D descriptor generation techniques.

Sub-Cellular Localization and Complex Formation by Aminoacyl-tRNA Synthetases in Cyanobacteria: Evidence for Interaction of Membrane-Anchored ValRS with ATP Synthase

tRNAs are charged with cognate amino acids by aminoacyl-tRNA synthetases (aaRSs) and subsequently delivered to the ribosome to be used as substrates for gene translation. Whether aminoacyl-tRNAs are channeled to the ribosome by transit within translational complexes that avoid their diffusion in the cytoplasm is a matter of intense investigation in organisms of the three domains of life. In the cyanobacterium Anabaena sp. PCC 7120, the valyl-tRNA synthetase (ValRS) is anchored to thylakoid membranes by means of the CAAD domain. We have investigated whether in this organism ValRS could act as a hub for the nucleation of a translational complex by attracting other aaRSs to the membranes. Out of the 20 aaRSs, only ValRS was found to localize in thylakoid membranes whereas the other enzymes occupied the soluble portion of the cytoplasm. To investigate the basis for this asymmetric distribution of aaRSs, a global search for proteins interacting with the 20 aaRSs was conducted. The interaction between ValRS and the FoF1 ATP synthase complex here reported is of utmost interest and suggests a functional link between elements of the gene translation and energy production machineries.

ViDRILO: The Visual and Depth Robot Indoor Localization with Objects information dataset

In this article we describe a semantic localization dataset for indoor environments named ViDRILO. The dataset provides five sequences of frames acquired with a mobile robot in two similar office buildings under different lighting conditions. Each frame consists of a point cloud representation of the scene and a perspective image. The frames in the dataset are annotated with the semantic category of the scene, but also with the presence or absence of a list of predefined objects appearing in the scene. In addition to the frames and annotations, the dataset is distributed with a set of tools for its use in both place classification and object recognition tasks. The large number of labeled frames in conjunction with the annotation scheme make this dataset different from existing ones. The ViDRILO dataset is released for use as a benchmark for different problems such as multimodal place classification and object recognition, 3D reconstruction or point cloud data compression.