Slanted Planes Research Articles

Semi-empirical terramechanics modelling of rough terrain represented by a height field

Dynamic simulations of various types of off-road vehicles, from planetary rovers to agricultural equipment, have long relied on well-established semi-empirical terramechanics models. While these models do have drawbacks and reliability issues that have been addressed by numerous works in the decades since the models were first introduced, semi-empirical approaches remain one of the few ways to simulate realistic wheel-soil interaction in real-time. One of their drawbacks is their assumption that the terrain is a flat plane. The models work by integrating normal and shear stresses along the wheel-terrain contact patch. The normal stress at each point along the contact patch is determined using an equation that computes soil pressure based on semi-empirical parameters, the dimensions of the wheel and the sinkage, which is determined based on the distance between the point and the plane that defines the terrain. Other works simplify the rough terrain contact problem by defining an equivalent contact plane at each time step in order to continue to be able to use semi-empirical models - modified to work with slanted planes - to compute the interaction forces. In this work, we propose a new, modified version of the semi-empirical model in which interaction forces for a wheel travelling on rough terrain can be computed without the need to use an equivalent contact plane. To highlight the advantages of our proposed approach, we compare our simulation results to the results of simulations using an existing approach for modelling a wheel travelling over rough terrain using traditional semi-empirical models.

Binocular viewing geometry shapes neural processing of slanted planes: Results from theoretical V1 modeling and human psychophysics

An important component of real-world scene perception is the estimation of 3D surface orientation (slant and tilt). However, neither the neural circuits that compute 3D orientation nor the supporting computations have been fully uncovered. To help address that gap, we apply a projective geometry framework to the study of both the neural processing and perception of 3D orientation, building on recent work that suggests the mapping of the 3D environment onto the two retinae shapes both the tuning of neurons in area MT to motion-in-depth and corresponding patterns of perceptual errors in human psychophysics (Bonnen et al., 2020). The projection of slanted planes results in interocular disparities in retinal orientation, and interestingly, binocular V1 neurons can have different monocular orientation preferences (Bridge & Cumming, 2001), which could make them sensitive to orientation disparities. Therefore, we have constructed an encoding model that predicts responses of binocular V1 neurons to the orientation disparities produced by a stimulus plane with varying slant, given each theoretical neuron’s monocular orientation preferences and ocular dominance level. The resulting slant tuning curves are noncanonical in shape, do not tile slant evenly, and are strongly affected by viewing distance. Notably, at greater viewing distances, extreme slants (beyond approximately +/- 70 degrees) are encoded by steep portions of the curves, but moderate and low slants are encoded by relatively flat portions. As expected based on those observations, our model population decoder performs best at extreme slants, and performance degrades as slant approaches zero and as viewing distance increases. We additionally observed these patterns in human perceptual performance on a slant discrimination task. These results support the theory that the neural processing of 3D scene components is shaped by projective geometry and represent a step toward advancing understanding of the computations supporting 3D orientation perception.

Recognizing Slanted Deck Scenes by Non-Manhattan Spatial Right Angle Projection

Recognizing slanted deck scenes is crucial to security monitoring for protecting ships and making them behave smartly resilient. However, there are multitude of diverse structures that are designed as slanted planes due to rough maritime environments. Traditional methods for scene understanding from 3D point clouds or RGB-D data are energy-consuming and memory intensive, which makes those models less reliable in a resource-constrained system of limited compute, memory, and energy resources on ships. In this study, we present an approach to understanding deck scenes, including slanted structures, using a low-cost monocular camera without prior training. New clusters of slanted angle projections are extracted. The vanishing points of slanted non-Manhattan angle projections are estimated. These slanted planes can be reshaped by compositions of non-Manhattan angle projections. Combined with Manhattan planes, a deck scene can be approximated by Manhattan and non-Manhattan planes. Unlike deep learning-based algorithms, this approach requires no prior training or knowledge of the camera’s internal parameters. Experimental results demonstrated that the method can successfully elucidate diverse elements, including slanted structures, meeting safety monitoring requirements using a resource-constrained monocular camera in a deck environment.

Understanding the cause of perceptual depth-sign ambiguity with dihedral angles created with motion parallax

Early research employing computer-generated dihedrals suggested that depth from motion parallax (MP) is unstable and depth-sign ambiguous. The pursuit theory of motion parallax suggests that perceptual instability and depth-sign reversals are the result of stimulus parameters exceeding the physically-possible conditions for a rigid stimulus as described by the Motion/Pursuit Law. When the Motion/Pursuit Ratio approaches 1, depth from MP approaches infinity. The visual system may recognize when stimulus retinal image velocity exceeds pursuit velocity (M/PR > 1,) representing an impossible rigid object. Such conditions are easily generated with computer-generated MP stimuli, producing depth-sign reversals and perhaps concomitant changes to perceived depth magnitude. To test this, we are measured perceived depth magnitude, in addition to depth-sign, with dihedral stimuli. Psychophysical stimuli included both physical and virtual dihedrals having two planes intersecting with a vertex on the horizontal meridian, facing or opposing the observer. With the 40 cm viewing distance, both stimulus types subtended 10.6 degrees. The virtual random-dot MP stimuli used M/PR gradients to depict the two slanted planes with varying relative depth magnitudes (60-110 mm) and therefore varying slants (19-30 degrees). Physical stimuli were 3D printed with varying slants (15-75 degrees). Observers indicated both slant of the front-facing plane using orientation of their palm, and perceived depth-sign of the vertex. With full-cue viewing of physical stimuli, observers underestimated slant by half (individuals varied between 0.3-0.7). Because an underestimate of slant signals an increase in perceived depth, the reported slants of MP stimuli were corrected with functions derived from the physical stimuli. Subsequently with MP stimuli, observers overestimated slopes indicating, similar to previous work, remarkable underestimation of perceived depth magnitude. In addition, increasing MP stimulus parameters to represent depth magnitudes that are impossible within the framework of the Motion/Pursuit Law produces both depth-sign reversals and reductions in perceived depth magnitude.

Full View Optical Flow Estimation Leveraged From Light Field Superpixel

In this paper, we present a full view optical flow estimation method for plenoptic imaging. Our method employs the structure delivered by the four-dimensional light field over multiple views making use of superpixels. These superpixels are four dimensional in nature and can be used to represent the objects in the scene as a set of slanted-planes in three-dimensional space so as to recover a piecewise rigid depth estimate. Taking advantage of these superpixels and the corresponding slanted planes, we recover the optical flow and depth maps by using a two-step optimization scheme where the flow is propagated from the central view to the other views in the imagery. We illustrate the utility of our method for depth and flow estimation making use of a dataset of synthetically generated image sequences and real-world imagery captured using a Lytro Illum camera. We also compare our results with those yielded by a number of alternatives elsewhere in the literature.

Finite Element Analysis of Creep Crack Initiation in Functionally Graded Materials with Crack Parallel to the Gradient

With the advances in material synthesis technologies, functionally graded materials (FGMs) are developed to use in high temperature structurals due to the excellent high temperature mechanical properties. To facilitate wide use of FGMs in high temperature structures, finite element method (FEM) was used in this paper to investigate effects of creep resistant properties gradients on creep crack initiation (CCI) in FGMs, with crack parallel to the gradient. Results indicated that when creep resistant properties increased in the crack growth direction, CCI was retarded by creep properties gradients. However, CCI was accelerated by creep properties gradients when creep resistant properties decreased in the crack growth direction. CCI position occurred in the two symmetric slanted planes of the initial crack, regardless of the gradient variation of creep resistant properties.

REDUCTION OF THE FRONTO-PARALLEL BIAS FOR WIDE-BASELINE SEMI-GLOBAL MATCHING

Abstract. Semi-Global Matching (SGM) is a widely-used technique for dense image matching that is popular because of its accuracy and speed. While it works well for textured scenes, it can fail on slanted surfaces particularly in wide-baseline configurations due to the so-called fronto-parallel bias. In this paper, we propose an extension of SGM that utilizes image warping to reduce the fronto-parallel bias in the data term, based on estimating dominant slanted planes. The latter are also used as surface priors improving the smoothness term. Our proposed method calculates disparity maps for each dominant slanted plane and fuses them to obtain the final disparity map. We have quantitatively evaluated our approach outperforming SGM and SGM-P on synthetic data and demonstrate its potential on real data by qualitative results. In this way, we underscore the need to tackle the fronto-parallel bias in particular for wide-baseline configurations in both the data term and the smoothness term of SGM.

Emergence of symmetry selectivity in the visual areas of the human brain: fMRI responses to symmetry presented in both frontoparallel and slanted planes

Symmetry is effortlessly perceived by humans across changes in viewing geometry. Here, we re‐examined the network subserving symmetry processing in the context of up‐to‐date retinotopic definitions of visual areas. Responses in object selective cortex, as defined by functional localizers, were also examined. We further examined responses to both frontoparallel and slanted symmetry while manipulating attention both toward and away from symmetry. Symmetry‐specific responses first emerge in V3 and continue across all downstream areas examined. Of the retinotopic areas, ventral occipital VO1 showed the strongest symmetry response, which was similar in magnitude to the responses observed in object selective cortex. Neural responses were found to increase with both the coherence and folds of symmetry. Compared to passive viewing, drawing attention to symmetry generally increased neural responses and the correspondence of these neural responses with psychophysical performance. Examining symmetry on the slanted plane found responses to again emerge in V3, continue through downstream visual cortex, and be strongest in VO1 and LOB. Both slanted and frontoparallel symmetry evoked similar activity when participants performed a symmetry‐related task. However, when a symmetry‐unrelated task was performed, fMRI responses to slanted symmetry were reduced relative to their frontoparallel counterparts. These task‐related changes provide a neural signature that suggests slant has to be computed ahead of symmetry being appropriately extracted, known as the “normalization” account of symmetry processing. Specifically, our results suggest that normalization occurs naturally when attention is directed toward symmetry and orientation, but becomes interrupted when attention is directed away from these features.

Microstructures and mechanical behavior of magnesium processed by ECAP at ice-water temperature

Magnesium of high purity is processed by equal channel angular pressing (ECAP) up to eight passes at the ice-water temperature, in which a core–shell-like structure is used. The core–shell-like structure consists of pure iron (Fe) of 1.5 mm in thickness as the shell and magnesium (Mg) as the core. The microstructure, texture and mechanical behavior of the ECAP-processed Mg are studied. The ECAP processing leads to the formation of fine and equiaxed grains of ~1.1 µm. The basal planes initially parallel to the extrusion direction evolve to slanted basal planes with the tilting angle in a range of 25°–45° to the extrusion direction. Increasing the number of the extrusion passes leads to the decreasing of twins and dislocation density in grains, while individual grains after eight passes still have high dislocation density. The large decreases of twins and the dislocation density make dynamic recrystallization (DRX) difficult, resulting in the decrease of the degree of DRX. Tension test reveals that the mechanical behavior of the ECAP-processed Mg is dependent on grain refinement and textures. The yield strength of the ECAP-extruded Mg first increases with the decrease of the grain size, and then decreases with further decrease of the grain size.

Planar structures with automatically generated bevel joints

A generative method based on computer algorithms is proposed to automatically produce parts of planar structures ready for fabrication. The parts resemble surface patches of the digital model of a structure. Each part is generated with bevel joints on the edges so that part-to-part connection is enabled by friction of the joints. The shape of a bevel joint is determined by the interior angle between two parts, and is modelled by a number of parameters, including the thickness of a planar material in use. The bevel joints consist of slanted planes, and in principle when they are assembled, no gap exists on the surface of the physical structure. Due to the slanted planes, the joints cannot be fabricated by a laser cutter that can only produce vertical cuts. We experimented the fabrication with a three-axis CNC router and a 3D printer; both produced accurate and robust parts; however, there is limitation in using a CNC cutter, which is discussed in details.

Direct Digital Surface Model Generation by Semi-Global Vertical Line Locus Matching

As the core issue for Digital Surface Model (DSM) generation, image matching is often implemented in photo space to get disparity or depth map. However, DSM is generated in object space with additional processes such as reference image selection, disparity maps fusion or depth maps merging, and interpolation. This difference between photo space and object space leads to process complexity and computation redundancy. We propose a direct DSM generation approach called the semi-global vertical line locus matching (SGVLL), to generate DSM with dense matching in the object space directly. First, we designed a cost function, robust to the pre-set elevation step and projection distortion, and detected occlusion during cost calculation to achieve a sound photo-consistency measurement. Then, we proposed an improved semi-global cost aggregation with guidance of true-orthophoto to obtain superior results at weak texture regions and slanted planes. The proposed method achieves performance very close to the state-of-the-art with less time consumption, which was experimentally evaluated and verified using nadir aerial images and reference data.

A fast Galerkin-based method for eigenfrequencies in acoustics of small rooms with slanted boundary planes

This paper proposes a Galerkin-type numerical algorithm for an efficient calculation of low eigenfrequencies for rectangular parallelepiped rooms with slanted boundary planes, in the frequency interval (0,200)Hz, with the volume V≤200m3. The main idea of the algorithm is to apply a system of Galerkin׳s basis functions which are orthogonal, after a certain change of variables, in a unit cube. As a result, the problem is reduced to a classical problem of the computational algebra about eigenvalues of a symmetric matrix. If applied to any parallelepiped of non-splayed geometry, the algorithm automatically gives the known classical modes. For rooms with the splayed planes, the sought mode frequencies can precisely be calculated in real time on a personal computer, by taking from 11 to 13 basis functions along each coordinate axis. Some particular examples are considered, in order to demonstrate the capability of the proposed algorithm, as well as its precision with a change of basis functions. It is also seen from the discussed examples that splaying the boundary planes can indeed make more uniform distribution of the low natural frequencies.

Finding Optimal Hydraulic-Fracture Angles in Productivity-Maximized Shale Well Design

This article, written by Special Publications Editor Adam Wilson, contains highlights of paper SPE 170965, “Optimal Hydraulic-Fracture Angle in Productivity-Maximized Shale Well Design,” by Nadav Sorek, SPE, Jose A. Moreno, Ryan Rice, Guofan Luo, and Christine Ehlig-Economides, SPE, Texas A&M University, prepared for the 2014 SPE Annual Technical Conference and Exhibition, Amsterdam, 27–29 October. The paper has not been peer reviewed. In general, hydraulic fractures propagate perpendicular to the horizontal-well axis whenever drilling is parallel to the minimum principal-stress plane. However, operators frequently drill horizontal wells parallel to lease boundaries, resulting in slanted hydraulic-fracture planes at angles less than 90° from the well axis. This study provides a model for the inclined fracture case. It applies and further extends the unified-fracture-design approach for rectangular drainage areas, relating the dimensionless proppant number to the maximum productivity index in pseudosteady-state conditions. Introduction Industry experience suggests that horizontal shale-gas development is enhanced by drilling in the direction parallel to the local minimum principal horizontal stress. Because US mineral leases frequently are rectangular areas with north/south and east/west boundaries, operators often drill parallel to the lease boundaries, prioritizing well saturation over optimum fracture-length propagation. This practice leads to the creation of hydraulic-fracture planes that are slanted at an angle less than 90° with respect to the well axis. Extensive research has investigated the effects of angled fractures on well productivity. Theoretical and empirical data have been used to address the effect of wellbore azimuth on well performance in the Marcellus shale. Results showed that, for each degree a well was suboptimal to minimum horizontal stress, the estimated ultimate recovery decreased by 7.25 Mscf per foot of effective lateral length. Unified-fracture-design methodology indicates the hydraulic- fracturing treatment design that maximizes well productivity for any set of reservoir and proppant properties and a given injected-proppant mass. This methodology introduces the concept of proppant number, which describes the weighted ratio of propped-fracture volume to a square reservoir volume. This study applies and further extends the unified-fracture-design approach to show that, for any given set of reservoir and proppant properties along with a given proppant mass, as long as the created fractures define the same stimulated rock volume, there exists a well direction resulting in maximized well productivity that is not parallel to the minimum-stress direction. First, a correlation is established between the proppant number and the optimum drainage-area aspect ratio. Then, this correlation is used to express what the optimum fracture angle is for a specific proppant number.

On the Spectrum of the Plenoptic Function.

The plenoptic function is a powerful tool to analyze the properties of multi-view image data sets. In particular, the understanding of the spectral properties of the plenoptic function is essential in many computer vision applications, including image-based rendering. In this paper, we derive for the first time an exact closed-form expression of the plenoptic spectrum of a slanted plane with finite width and use this expression as the elementary building block to derive the plenoptic spectrum of more sophisticated scenes. This is achieved by approximating the geometry of the scene with a set of slanted planes and evaluating the closed-form expression for each plane in the set. We then use this closed-form expression to revisit uniform plenoptic sampling. In this context, we derive a new Nyquist rate for the plenoptic sampling of a slanted plane and a new reconstruction filter. Through numerical simulations, on both real and synthetic scenes, we show that the new filter outperforms alternative existing filters.

Outdoor Map Construction Based on Aerial Photography and Electrical Map Using Multi-Plane Laser Range Scan Data

This research is on personal mobility that estimates its self position on a sensor data map created from sensor data, acquired from laser range scan sensors and/or other sensors, and annotates various multiple items of information on a digital map. This paper describes a method of creating an edge-based grid map from both aerial photography and an electricalmap for this purpose and a way and its realization to estimate position and to construct outdoor maps from multi-plane laser range scan data on the grid map. Since threedimensional scanning is rather difficult and the scan rate is low, we used two-dimensional scanning that enables movement without slowing it down by scanning multiple horizontal and/or slanted planes. Experimental results show that the system is able to ensure the accuracy of accumulated error within 2 m by integrating aerial photography and electrical maps plus multiplane scanning.

Depth as a latent variable

Event Abstract Back to Event Depth as a latent variable Christian Conrad1 and Roland Memisevic1* 1 University of Frankfurt, Germany Estimation of binocular disparity in the brain is widely assumed to be based on the comparison of local phase information from binocular receptive fields. The classic binocular energy model shows that this requires the presence of receptive fields in the form of local quadrature pairs within the eye and with phase- or position-shifts across the eyes. While numerous computational accounts of stereopsis have been based on these observations, there has been little work on how energy-models can emerge through learning from the statistics of image pairs. We describe a way to cast binocular disparity estimation as a probabilistic model, and we show how learning on data-bases of multi-camera views of a scene leads to position-shifted Gabor filters. The model agrees with the classical binocular energy and cross-correlation models in that it learns shifted quadrature pairs, while also allowing for more flexible connectivity that yields richer dependencies between the learned filters. Learning on binocular data with known ground-truth disparities furthermore makes it possible to train a network to perform depth estimation entirely based on training data. In contrast to standard (monocular) feature learning, acquiring training data for binocular feature learning is more challenging as it requires the generation of image pairs in accordance with a given camera setup. We use a methodology for generating image pairs and corresponding ground truth disparities based on recent evidence that due to fixations and vergence, biological receptive fields are confronted mainly with locally smooth surfaces. To generate the training data, we first generate depth maps as slanted planes in 3D. We then generate texture maps using patches cropped from natural images of the same size as the depth maps. Finally, we project the 3D scene onto a set of cameras defined by their camera matrices. We demonstrate how the presence of ground truth disparities makes it possible to learn depth estimation, by adding a pooling-layer that is trained to predict the ground-truth disparities from the inferred binocular encoding. We demonstrate this purely learning-based depth estimation scheme on random-dot stereograms. Keywords: neural coding and decoding, stereopsis Conference: BC11 : Computational Neuroscience & Neurotechnology Bernstein Conference & Neurex Annual Meeting 2011, Freiburg, Germany, 4 Oct - 6 Oct, 2011. Presentation Type: Poster Topic: neural encoding and decoding (please use "neural coding and decoding" as keyword) Citation: Conrad C and Memisevic R (2011). Depth as a latent variable. Front. Comput. Neurosci. Conference Abstract: BC11 : Computational Neuroscience & Neurotechnology Bernstein Conference & Neurex Annual Meeting 2011. doi: 10.3389/conf.fncom.2011.53.00179 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 22 Aug 2011; Published Online: 04 Oct 2011. * Correspondence: Prof. Roland Memisevic, University of Frankfurt, Frankfurt, 60325, Germany, ro@cs.uni-frankfurt.de Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Christian Conrad Roland Memisevic Google Christian Conrad Roland Memisevic Google Scholar Christian Conrad Roland Memisevic PubMed Christian Conrad Roland Memisevic Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

The Selectivity of Neurons in the Macaque Fundus of the Superior Temporal Area for Three-Dimensional Structure from Motion

Motion is a potent cue for the perception of three-dimensional (3D) shape in primates, but little is known about its underlying neural mechanisms. Guided by recent functional magnetic resonance imaging results, we tested neurons in the fundus of the superior temporal sulcus (FST) area of two macaque monkeys (Macaca mulatta, one male) using motion-defined surface patches with various 3D shapes such as slanted planes, saddles, or cylinders. The majority of the FST neurons (>80%) were selective for stimuli depicting specific shapes, and all the surfaces tested were represented among the selective FST neurons. Importantly, this selectivity tolerated changes in speed, position, size, or between binocular and monocular presentations. This tolerance demonstrates that the 3D structure-from-motion (3D-SFM) selectivity of FST neurons is a higher-order selectivity, which cannot be reduced to a lower-order speed selectivity. The 3D-SFM selectivity of FST neurons was unaffected by removal of the opposed-motion cue that supplemented the speed gradient cue in the standard stimuli. When tested with the same standard stimuli, fewer neurons in the middle temporal/visual 5 (MT/V5) area were selective than FST neurons. In addition, selective MT/V5 neurons represented fewer types of surfaces and were less tolerant of stimulus changes than FST neurons. Overall, these results indicate that FST neurons code motion-defined 3D shape fragments, underscoring the central role of FST in processing 3D-SFM.

Oblique-Directional Plasma Etching of Si Using a Faraday Cage

A method for controlling the direction of Si etch profiles was developed using a Faraday cage system and a conventional gas-chopping process. The gas-chopping process consisted of sequentially alternating etching and deposition steps using and plasmas, respectively. The single-directional oblique etching was achieved using a Faraday cage having a top horizontal grid plane, inside of which a tilted substrate holder was placed. The double-directional etching was also obtained using a cage having two slanted grid planes. The double-directional etch profiles were either symmetrically or asymmetrically oblique according to the angle of the sample holder. During the oblique-directional etching, the shadowing effect by the bombardment of tilt-angled ions was found to affect etch profiles. Due to the shadowing effect, the top width of the pattern and the etch depth decreased as the ion-incident angle increased, and a fluorocarbon film accumulated in the shadow area. By regulating the process conditions to allow a higher mask selectivity, an oblique Si etch profile with a high aspect ratio (etch and aspect ) was obtained.

3D shape perception from combined depth cues in human visual cortex

Our perception of the world's three-dimensional (3D) structure is critical for object recognition, navigation and planning actions. To accomplish this, the brain combines different types of visual information about depth structure, but at present, the neural architecture mediating this combination remains largely unknown. Here, we report neuroimaging correlates of human 3D shape perception from the combination of two depth cues. We measured fMRI responses while observers judged the 3D structure of two sequentially presented images of slanted planes defined by binocular disparity and perspective. We compared the behavioral and fMRI responses evoked by changes in one or both of the depth cues. fMRI responses in extrastriate areas (hMT+/V5 and lateral occipital complex), rather than responses in early retinotopic areas, reflected differences in perceived 3D shape, suggesting 'combined-cue' representations in higher visual areas. These findings provide insight into the neural circuits engaged when the human brain combines different information sources for unified 3D visual perception.

Estimating the perspective pose of texture planes using spectral analysis on the unit sphere

Multiple vanishing point detection provides the key to recovering the perspective pose of textured planes. If vanishing points are to be detected from spectral information then there are two computational problems that need to be solved. Firstly, the search of the extended image plane is unbounded, and hence the location of vanishing points at or near infinity is difficult. Secondly, correspondences between local spectra need to be established so that vanishing points can be triangulated. In this paper, we offer a way of overcoming these two difficulties. We overcome the problem of unbounded search by mapping the information provided by local spectral moments onto the unit sphere. According to our representation, the position and direction of each local spectrum maps onto a great circle on the unit sphere. The need for correspondences is overcome by accumulating the great circle intercepts. Vanishing points occur at local accumulator maxima on the unit sphere. To improve the accuracy of the recovered perspective pose parameters for highly slanted planes, we use an adaptive spectral window. This selects the window so as to reduce spectral defocusing by minimising the determinant of the spectral covariance matrix. We experiment with the new shape-from-texture technique on both synthetic and real world data. Here it proves to be an accurate and robust means of estimating perspective pose.