The use of machine learning methods to tackle challenging physical layer signal processing tasks has attracted significant attention. In this work, we focus on the use of neural networks (NNs) to perform pilot-assisted channel estimation in an OFDM system in order to avoid the challenging task of estimating the channel covariance matrix. In particular, we perform a systematic design-space exploration of NN configurations, quantization, and pruning in order to improve feedforward NN architectures that are typically used in the literature for the channel estimation task. We show that choosing an appropriate NN architecture is crucial to reduce the complexity of NN-assisted channel estimation methods. Moreover, we demonstrate that, similarly to other applications and domains, careful quantization and pruning can lead to significant complexity reduction with a negligible performance degradation. Finally, we show that using a solution with multiple distinct NNs trained for different signal-to-noise ratios interestingly leads to lower overall computational complexity and storage requirements, while achieving a better performance with respect to using a single NN trained for the entire SNR range.

Artificial Neural Networks (NNs) can effectively be used to solve many classification and regression problems, and deliver state-of-the-art performance in the application domains of natural language processing (NLP) and computer vision (CV). However, the tremendous amount of data movement and excessive convolutional workload of these networks hampers large-scale mobile and embedded productization. Therefore these models are generally mapped to energy-efficient accelerators without floating-point support. Weight and data quantization is an effective way to deploy high-precision models to efficient integer-based platforms. In this paper a quantization method for platforms without wide accumulation registers is being proposed. Two constraints to maximize the bit width of weights and input data for a given accumulator size are introduced. These constraints exploit knowledge about the weight and data distribution of individual layers. Using these constraints, we propose a layer-wise quantization heuristic to find a good fixed-point network approximation. To reduce the number of configurations to consider, only solutions that fully utilize the available accumulator bits are being tested. We demonstrate that 16-bit accumulators are able to obtain a Top-1 classification accuracy within 1% of the floating-point baselines on the CIFAR-10 and ILSVRC2012 image classification benchmarks.

Inexpensive user tracking is an important problem in various application domains such as healthcare, human-computer interaction, energy savings, safety, robotics, security and so on. Yet, it cannot be easily solved due to its probabilistic nature, high level of abstraction and uncertainties, on the one side, and to the limitations of our current technologies and learning algorithms, on the other side. In this paper, we tackle this problem by using the Multi-integrated Sensor Technology, which comes at a low price. At the same time, we are aiming to address the lightweight learning requirements by investigating Factored Conditional Restricted Boltzmann Machines (FCRBMs), a form of Deep Learning, that has proven to be an efficient and effective machine learning framework. However, due to their construction properties, the conventional FCRBMs are only capable of performing predictions but are not capable of making classification. Herein, we are proposing extended FCRBMs (eFCRBMs), which incorporate a novel classification scheme, to solve this problem. Experiments performed on both artificially generated as well as real-world data demonstrate the effectiveness and efficiency of the proposed technique. We show that eFCRBMs outperform popular approaches including Support Vector Machines, Naive Bayes, AdaBoost, and Gaussian Mixture Models.

An annotated data set is presented meant to help researchers in developing, evaluating, and comparing various approaches in robotics for building space representations appropriate for communicating with humans. The data consist of omnidirectional images, laser range scans, sonar readings, and robot odometry. A set of base-level human spatial concepts is used to annotate the data.

The ”layered image model ” [13] represents an image sequence as a composition of 2D layers where each layer corresponds to a different object. A layer is described by its appearance and its transparency mask. The transparency masks are used to combine the layers. In this paper we present a probabilistic layered model that uses the ”logistic principal component analysis (PCA) ” to describe the masks. The Gaussian based factor analysis was used previously but it does not consider the constraints imposed on the transparency values. The ”logistic PCA ” models the transparency values that are between 0 and 1 more naturally using Bernoulli distributions. The presented model can be used to automatically extract low dimensional representation of the transparency maps of the moving objects from a video sequences more efficiently. 1

Vision systems are used more and more in 'personal' robots interacting with humans, since semantic information about objects and places can be derived from the rich sensory information. Visual information is also used for building appearance based topological maps, which can be used for localization. In this paper we describe a system capable of using this appearance based topological map for navigation. The system is made robust by using the epipolar geometry and a planar floor constraint in computing the necessary heading information. Using this method the robot is able to drive robustly in a large environment. We tested the method on real data under varying environment conditions and compared performance with a human-controlled robot.

This paper addresses the problem of people detection using 2D range data and omnidirectional vision. The range data is often used in robotics to detect person legs. First, we will develop a reliable 2D range data leg detector using the AdaBoost algorithm. Second, instead of using the leg detector to detect people, we propose a more reliable method that takes into account the spatial arrangement of the detected legs. The method is inspired by the latest results on the "part-based representations" from the computer vision area. Finally, we show how the part-based representations can be used to combine the leg detections from the 2D range data and upper body, lower body and full body detections from omnidirectional camera. The experimental results show highly reliable people detection when the two sensors are combined. We use Haar-like features and the AdaBoost again to construct the omnicam body parts detectors.

There are various situations where image data is binary: character recognition, result of image segmentation etc. As a first contribution, we compare Gaussian based principal component analysis (PCA), which is often used to model images, and "binary PCA" which models the binary data more naturally using Bernoulli distributions. Furthermore, we address the problem of data alignment. Image data is often perturbed by some global transformations such as shifting, rotation, scaling etc. In such cases the data needs to be transformed to some canonical aligned form. As a second contribution, we extend the binary PCA to the "transformation invariant mixture of binary PCAs" which simultaneously corrects the data for a set of global transformations and learns the binary PCA model on the aligned data.

Mobile robot localization and navigation requires a map - the robot's internal representation of the environment. A common problem is that path planning becomes very inefficient for large maps. In this paper we address the problem of segmenting a base-level map in order to construct a higher-level representation of the space which can be used for more efficient planning. We represent the base-level map as a graph for both geometric and appearance based space representations. Then we use a graph partitioning method to cluster nodes of the base-level map and in this way construct a high-level map, which is also a graph. We apply a hierarchical path planning method for stochastic tasks based on Markov decision processes (MDPs) and investigate the effect of choosing different numbers of clusters

In appearance based robot localization a new image is matched with every image in the database. In this paper we describe how to reduce the number of images in this database with minimal loss of information and thereby increasing the efficiency of localization significantly. First we build an appearance based model consisting of a graph in which the nodes denote images and links denote relations between images. This graph is then pruned using the connected dominating set algorithm. The method is tested on an image set acquired by a mobile robot equipped with an omnidirectional camera driving around in an office environment, as well as sets of images taken while driving through a real furnished home environment

An ambitious goal in modern robotic science is to build mobile robots that are able to interact as companions in real world environments. Especially for caretaking of elderly people a system robustly working at private homes is essential, requiring a very natural and human oriented way of communication. Since home environments are usually very individual a first task for a newly acquired robot is to get familiar with its new environment. This paper gives a short overview on how we integrated a vision based localization using the advantages of a very modular architecture and extending a spoken dialog system for online labeling and interaction about different locations. We present results from the integrated system working in a real, fully furnished home environment where it was able to learn the names of different rooms. This system enables us to perform real user studies in future without the need to fall back to Wizard-of-Oz experiments. Ongoing work aims at enabling the robot to take initiative by asking for unknown locations. A future extension is the ability to generalize over features of known rooms to make predictions when encountering unknown rooms

When determining subject preferences for human-robot interaction, an important issue is the interpretation of the subjects' responses during the trials. Employing a non-intrusive approach, this paper discusses the methodological issues for annotating vision data by allowing the subjects to indicate their comfort using a handheld comfort level device during the trials. In previous research, the analysis of collected comfort and vision data was made difficult due to problems concerning the manual synchronization of different modalities. In the current paper, we overcome this issue by real-time integration of the subject's feedback on subjective comfort into the video stream. The implications for more efficient analysis of human-robot interaction data, as well as possible future developments of this approach are discussed

This paper shows that the common task of interest region matching using local descriptors can be improved using a new similarity measure. The similarity measure is motivated by the information theoretic image alignment that maximize mutual information between images. A property of the mutual information metric is that it does not only depend on how similar the signals are but also how complex they are. We present how similar logic can be applied to the standard SIFT descriptor. The results show improvement at almost no additional computational costs.

In this paper we describe a system that enables a mobile robot equipped with a color vision system to track humans in indoor environments. We developed a method for tracking humans when they are within the field of view of the camera, based on motion and color cues. However, the robot also has to keep track of humans which leave the field of view and re-enter later. We developed a dynamic Bayesian network for such a global tracking task. Experimental results on real data confirm the viability of the developed method.

This paper addresses the question how robot planning (e.g. for navigation) can be done with hierarchical maps. We present an algorithm for hierarchical path planning for stochastic tasks, based on Markov decision processes (MDPs) and dynamic programming. It is more efficient than standard dynamic programming for "flat" MDPs, because it reduces the state space for all levels in its hierarchy and it allows reuse of previously computed partial policies. This computational advantage comes at the cost of some extra memory and overhead to represent and coordinate the hierarchical system, and in some cases somewhat longer paths to target locations. We demonstrate the method on artificially generated MDP data, and on real robot data from our vision-controlled robot navigating in an office environment.

This paper addresses the problem of automatic construction of a hierarchical map from images. Our approach departs from a large collection of omnidirectional images taken at many locations in a building. First, a low-level map is built that consists of a graph in which relations between images are represented. For this, we use a metric based on visual landmarks (SIFT features) and geometrical constraints. Then, we use a graph partitioning method to cluster nodes and in this way construct the high-level map. Experiments on real data show that meaningful higher and lower level maps are obtained, which can be used for accurate localization and planning.

There are two open problems when finite mixture densities are used to model multivariate data: the selection of the number of components and the initialization. In this paper, we propose an online (recursive) algorithm that estimates the parameters of the mixture and that simultaneously selects the number of components. The new algorithm starts with a large number of randomly initialized components. A prior is used as a bias for maximally structured models. A stochastic approximation recursive learning algorithm is proposed to search for the maximum a posteriori (MAP) solution and to discard the irrelevant components.