Matching Recall and Storage in Sequence Learning with Spiking Neural Networks

Storing and recalling spiking sequences is a general problem the brain needs to solve. It is, however, unclear what type of biologically plausible learning rule is suited to learn a wide class of spatiotemporal activity patterns in a robust way. Here we consider a recurrent network of stochastic spiking neurons composed of both visible and hidden neurons. We derive a generic learning rule that is matched to the neural dynamics by minimizing an upper bound on the Kullback–Leibler divergence from the target distribution to the model distribution. The derived learning rule is consistent with spike-timing dependent plasticity in that a presynaptic spike preceding a postsynaptic spike elicits potentiation while otherwise depression emerges. Furthermore, the learning rule for synapses that target visible neurons can be matched to the recently proposed voltage-triplet rule. The learning rule for synapses that target hidden neurons is modulated by a global factor, which shares properties with astrocytes and gives rise to testable predictions.

A Closed-Form, Consistent and Robust Solution to Uncalibrated Photometric Stereo Via Local Diffuse Reflectance Maxima

Images of an object under different illumination are known to provide strong cues about the object surface. A mathematical formalization of how to recover the normal map of such a surface leads to the so-called uncalibrated photometric stereo problem. In the simplest instance, this problem can be reduced to the task of identifying only three parameters: the so-called generalized bas-relief (GBR) ambiguity. The challenge is to find additional general assumptions about the object, that identify these parameters uniquely. Current approaches are not consistent, i.e., they provide different solutions when run multiple times on the same data. To address this limitation, we propose exploiting local diffuse reflectance (LDR) maxima, i.e., points in the scene where the normal vector is parallel to the illumination direction (see Fig. 1). We demonstrate several noteworthy properties of these maxima: a closed-form solution, computational efficiency and GBR consistency. An LDR maximum yields a simple closed-form solution corresponding to a semi-circle in the GBR parameters space (see Fig. 2); because as few as two diffuse maxima in different images identify a unique solution, the identification of the GBR parameters can be achieved very efficiently; finally, the algorithm is consistent as it always returns the same solution given the same data. Our algorithm is also remarkably robust: It can obtain an accurate estimate of the GBR parameters even with extremely high levels of outliers in the detected maxima (up to 80 % of the observations). The method is validated on real data and achieves state-of-the-art results.

Rapid Volume Loss from Two East Greenland Outlet Glaciers Quantified Using Repeat Stereo Satellite Imagery

The coastal portions of Kangerdlugssuaq and Helheim glaciers in southeast Greenland lost at least 51 +/- 8 km(-3) yr(-1) of ice between 2001-2006 due to thinning and retreat, according to an analysis of sequential digital elevation models (DEMs) derived from stereo ASTER satellite imagery. The dominant contribution to this ice loss was dynamic thinning caused by the acceleration in flow of both glaciers. Peak rates of change, including thinning rates of similar to 90 m yr(-1), coincided with the rapid increases in flow speed. Extrapolation of the measured data to the ice divides yields an estimated combined catchment volume loss of similar to 122 +/- 30 km(-3) yr(-1), which accounts for half the total mass loss from the ice sheet reported in recent studies. These catchment-wide volume losses contributed similar to 0.31 +/- 0.07 mm yr(-1) to global sea level rise over the 5-year observation period with the coastal regions alone contributing at least 0.1 +/- 0.02 mm yr(-1).

Optimizing stereo-to-multiview conversion for autostereoscopic displays

We present a novel stereo-to-multiview video conversion method for glasses-free multiview displays. Different from previous stereo-to-multiview approaches, our mapping algorithm utilizes the limited depth range of autostereoscopic displays optimally and strives to preserve the scene's artistic composition and perceived depth even under strong depth compression. We first present an investigation of how perceived image quality relates to spatial frequency and disparity. The outcome of this study is utilized in a two-step mapping algorithm, where we (i) compress the scene depth using a non-linear global function to the depth range of an autostereoscopic display and (ii) enhance the depth gradients of salient objects to restore the perceived depth and salient scene structure. Finally, an adapted image domain warping algorithm is proposed to generate the multiview output, which enables overall disparity range extension.

Uncalibrated Near-Light Photometric Stereo

In this work we solve the uncalibrated photometric stereo problem with lights placed near the scene. We investigate different image formation models and find the one that best fits our observations. Although the devised model is more complex than its far-light counterpart, we show that under a global linear ambiguity the reconstruction is possible up to a rotation and scaling, which can be easily fixed. We also propose a solution for reconstructing the normal map, the albedo, the light positions and the light intensities of a scene given only a sequence of near-light images. This is done in an alternating minimization framework which first estimates both the normals and the albedo, and then the light positions and intensities. We validate our method on real world experiments and show that a near-light model leads to a significant improvement in the surface reconstruction compared to the classic distant illumination case.

Computational Tools for Stereo and Light Field Photography

This thesis covers a broad part of the field of computational photography, including video stabilization and image warping techniques, introductions to light field photography and the conversion of monocular images and videos into stereoscopic 3D content. We present a user assisted technique for stereoscopic 3D conversion from 2D images. Our approach exploits the geometric structure of perspective images including vanishing points. We allow a user to indicate lines, planes, and vanishing points in the input image, and directly employ these as guides of an image warp that produces a stereo image pair. Our method is most suitable for scenes with large scale structures such as buildings and is able to skip the step of constructing a depth map. Further, we propose a method to acquire 3D light fields using a hand-held camera, and describe several computational photography applications facilitated by our approach. As the input we take an image sequence from a camera translating along an approximately linear path with limited camera rotations. Users can acquire such data easily in a few seconds by moving a hand-held camera. We convert the input into a regularly sampled 3D light field by resampling and aligning them in the spatio-temporal domain. We also present a novel technique for high-quality disparity estimation from light fields. Finally, we show applications including digital refocusing and synthetic aperture blur, foreground removal, selective colorization, and others.

Shape model, reference system definition, and cartographic mapping standards for comet 67P/Churyumov-Gerasimenko Stereo-photogrammetric analysis of Rosetta/OSIRIS image data

We analyzed more than 200 OSIRIS NAC images with a pixel scale of 0.9-2.4 m/pixel of comet 67P/Churyumov-Gerasimenko (67P) that have been acquired from onboard the Rosetta spacecraft in August and September 2014 using stereo-photogrammetric methods (SPG). We derived improved spacecraft position and pointing data for the OSIRIS images and a high-resolution shape model that consists of about 16 million facets (2 m horizontal sampling) and a typical vertical accuracy at the decimeter scale. From this model, we derive a volume for the northern hemisphere of 9.35 km(3) +/- 0.1 km(3). With the assumption of a homogeneous density distribution and taking into account the current uncertainty of the position of the comet's center-of-mass, we extrapolated this value to an overall volume of 18.7 km(3) +/- 1.2 km(3), and, with a current best estimate of 1.0 X 10(13) kg for the mass, we derive a bulk density of 535 kg/m(3) +/- 35 kg/m(3). Furthermore, we used SPG methods to analyze the rotational elements of 67P. The rotational period for August and September 2014 was determined to be 12.4041 +/- 0.0004 h. For the orientation of the rotational axis (z-axis of the body-fixed reference frame) we derived a precession model with a half-cone angle of 0.14 degrees, a cone center position at 69.54 degrees/64.11 degrees (RA/Dec J2000 equatorial coordinates), and a precession period of 10.7 days. For the definition of zero longitude (x-axis orientation), we finally selected the boulder-like Cheops feature on the big lobe of 67P and fixed its spherical coordinates to 142.35 degrees right-hand-rule eastern longitude and -0.28 degrees latitude. This completes the definition of the new Cheops reference frame for 67P. Finally, we defined cartographic mapping standards for common use and combined analyses of scientific results that have been obtained not only within the OSIRIS team, but also within other groups of the Rosetta mission.

Matching Social and Ecological Systems in Complex Ocean Fisheries

This paper considers ocean fisheries as complex adaptive systems and addresses the question of how human institutions might be best matched to their structure and function. Ocean ecosystems operate at multiple scales, but the management of fisheries tends to be aimed at a single species considered at a single broad scale. The paper argues that this mismatch of ecological and management scale makes it difficult to address the fine-scale aspects of ocean ecosystems, and leads to fishing rights and strategies that tend to erode the underlying structure of populations and the system itself. A successful transition to ecosystem-based management will require institutions better able to economize on the acquisition of feedback about the impact of human activities. This is likely to be achieved by multiscale institutions whose organization mirrors the spatial organization of the ecosystem and whose communications occur through a polycentric network. Better feedback will allow the exploration of fine-scale science and the employment of fine-scale fishing restraints, better adapted to the behavior of fish and habitat. The scale and scope of individual fishing rights also needs to be congruent with the spatial structure of the ecosystem. Place-based rights can be expected to create a longer private planning horizon as well as stronger incentives for the private and public acquisition of system relevant knowledge.

Preference Structure and Random Paths to Stability in Matching Markets

This paper examines how preference correlation and intercorrelation combine to influence the length of a decentralized matching market's path to stability. In simulated experiments, marriage markets with various preference specifications begin at an arbitrary matching of couples and proceed toward stability via the random mechanism proposed by Roth and Vande Vate (1990). The results of these experiments reveal that fundamental preference characteristics are critical in predicting how long the market will take to reach a stable matching. In particular, intercorrelation and correlation are shown to have an exponential impact on the number of blocking pairs that must be randomly satisfied before stability is attained. The magnitude of the impact is dramatically different, however, depending on whether preferences are positively or negatively intercorrelated.