Pair Of Frames Research Articles

Quality Threshold Clustering of Molecular Dynamics: A Word of Caution

Clustering Molecular Dynamics trajectories is a common analysis that allows grouping together similar conformations. Several algorithms have been designed and optimized to perform this routine task, and among them, Quality Threshold stands as a very attractive option. This algorithm guarantees that in retrieved clusters no pair of frames will have a similarity value greater than a specified threshold, and hence, a set of strongly correlated frames are obtained for each cluster. In this work, it is shown that various commonly used software implementations are flawed by confusing Quality Threshold with another simplistic well-known clustering algorithm published by Daura et al. (Daura, X.; van Gunsteren, W. F.; Jaun, B.; Mark, A. E.; Gademann, K.; Seebach, D. Peptide Folding: When Simulation Meets Experiment. Angew. Chemie Int. Ed. 1999, 38 (1/2), 236-240). Daura's algorithm does not impose any quality threshold for the frames contained in retrieved clusters, bringing unrelated structural configurations altogether. The advantages of using Quality Threshold whenever possible to explore Molecular Dynamic trajectories is exemplified. An in-house implementation of the original Quality Threshold algorithm has been developed in order to illustrate our comments, and its code is freely available for further use by the scientific community.

An Out-of-Plane Operated Soft Engine Driving Stretchable Zone Plate for Adjusting Focal Point of an Ultrasonic Beam.

This paper presents a soft engine which performs up-and-down motion with four planar film-structured ionic polymer—metal composites (IPMC) actuators. This soft engine assembled with a stretchable Fresnel zone plate is capable of tuning the focus of ultrasonic beam. Instead of conventional clamps, we employ 3D printed frame pairs with magnets and a conductive gold cloth to provide an alternative solution for securing the IPMC actuators during assembly. The design and analysis of the zone plate are carefully performed. The zone plate allows the plane ultrasonic wave to be effectively focused. The motion of IPMC actuators stretch the metal-foil-made zone plate to tune the focal range of the ultrasonic beam. The zone plate, 3D frames and IPMC actuators were fabricated, assembled and tested. The stiffness normal to the stretchable zone plate with varied designs was investigated and the seven-zone design was selected for our experimental study. The force responsible for clamping the IPMC actuators, controlled by the magnetic attraction between the fabricated frames, was also examined. The driving voltage, current and resulting displacement of IPMC actuation were characterized. The developed soft engine stretching the zone plate to tune the focal point of the ultrasonic beam up to 10% was successfully demonstrated.

Self-Supervised Optical Flow Estimation by Projective Bootstrap

Dense optical flow estimation is complex and time consuming, with state-of-the-art methods relying either on large synthetic data sets or on pipelines requiring up to a few minutes per frame pair. In this paper, we address the problem of optical flow estimation in the automotive scenario in a self-supervised manner. We argue that optical flow can be cast as a geometrical warping between two successive video frames and devise a deep architecture to estimate such transformation in two stages. First, a dense pixel-level flow is computed with a projective bootstrap on rigid surfaces. We show how such global transformation can be approximated with a homography and extend spatial transformer layers so that they can be employed to compute the flow field implied by such transformation. Subsequently, we refine the prediction by feeding a second, deeper network that accounts for moving objects. A final reconstruction loss compares the warping of frame X <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">t</sub> with the subsequent frame X <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">t+1</sub> and guides both estimates. The model has the speed advantages of end-to-end deep architectures while achieving competitive performances, both outperforming recent unsupervised methods and showing good generalization capabilities on new automotive data sets.

UASOL, a large-scale high-resolution outdoor stereo dataset

In this paper, we propose a new dataset for outdoor depth estimation from single and stereo RGB images. The dataset was acquired from the point of view of a pedestrian. Currently, the most novel approaches take advantage of deep learning-based techniques, which have proven to outperform traditional state-of-the-art computer vision methods. Nonetheless, these methods require large amounts of reliable ground-truth data. Despite there already existing several datasets that could be used for depth estimation, almost none of them are outdoor-oriented from an egocentric point of view. Our dataset introduces a large number of high-definition pairs of color frames and corresponding depth maps from a human perspective. In addition, the proposed dataset also features human interaction and great variability of data, as shown in this work.

A real-time 3D video analyzer for enhanced 3D audio–visual systems

With the recent advent of three-dimensional (3D) sound home theater systems (HTS), more and more TV viewers are experiencing rich, immersive auditory presence at home. In this paper, visual processing approaches are provided to make 3D audio–visual (AV) systems more realistic to the viewers. In the proposed system, a visual engine processes stereo video streams to extract a disparity map for each pair of left and right video frames. Then, the engine determines the video depth level representing each frame of the disparity map. An audio engine then gives 3D sound depth effects according to the estimated video depth, thereby making viewers’ audio–visual experiences more synchronized. Two video processing algorithms are devised to extract the video depth from each frame: one is based on object segmentation, which turns out to be too complex to be implemented in the field-programmable gate array (FPGA) employed for real-time processing; the other algorithm uses a much simpler histogram-based approach to determine the depth of each video frame, and hence, it is more suitable for FPGA implementations. Subjective listening test results support the effectiveness of the proposed approaches.

The Reality of Conflict Frames in Media: A Comparative Analysis of News Stories in The Hindu and Dawn with Reference to the Aftermath of Uri Attack

The role of media in conflict situations is vital. When it comes to reporting a news story especially a conflict, it has been an issue of discussion as media have all the powers to decide how the story has to be framed in reality. The paper substantiates the occurrence of a pair of conflict frames that are often used by media while reporting a conflict. A qualitative content analysis of post Uri attack publications in The Hindu (Indian daily) and Dawn (Pakistani Daily) for one month is considered for analysis. The predominant appearance of ‘Indicative’ and ‘Dispute’ Frames often in pair and their concurrent relation in both the newspapers manifest the abstract, concrete and neutral characteristics of the frames. These identified frames are capable enough to create understandings, emotions, cognitions and perceptions which in reality form opinions, perspectives and discussions. This study significantly focuses on the role of the media frames in reporting Indo-Pak Conflicts in a real world scenario that later on provides the major context of discussion and bilateral relations between the countries. When indicative frames are colluded in conflict reports, their destructive power is much more as the indications are often rooted in a way to escalate the dispute that occurred. The effectiveness and the intensity of these notable paired frames are discussed in the paper to comprehend the reality of conflict frames used by media.

Sums of A Pair of Orthogonal Frames

Frames are more stable as compared to bases under the action of a bounded linear operator. Sums of different frames under the action of a bounded linear operator are studied with the help of analysis, synthesis and frame operators. A simple construction of frames from the existing ones under the action of such an operator is presented here. It is shown that a frame can be added to its alternate dual frames, yielding a new frame. It is also shown that the sum of a pair of orthogonal frames is a frame. This provides an easy construction of a frame where the frame bounds can be computed easily. Moreover, for a pair of orthogonal frames, the necessary and sufficient condition is presented for their alternate dual frames to be orthogonal. This allows for an easy construction of a large number of new frames.

Tight frames, Hadamard matrices and Zauner’s conjecture

We show that naturally associated to a SIC (symmetric informationally complete positive operator valued measure or SIC-POVM) in dimension d there are a number of higher dimensional structures: specifically a projector and complex Hadamard matrix in dimension d2, and a pair of ETFs (equiangular tight frames) in dimensions . We also show that a WH (Weyl–Heisenberg covariant) SIC in odd dimension d is naturally associated to a pair of symmetric tight fusion frames in dimension d. We deduce two relaxations of the WH SIC existence problem. We also find a reformulation of the problem in which the number of equations is fewer than the number of variables. Finally, we show that in at least four cases the structures associated to a SIC lie on continuous manifolds of such structures. In two of these cases the manifolds are non-linear. Restricted defect calculations are consistent with this being true for the structures associated to every known SIC with d between 3 and 16, suggesting it may be true for all .

P-woven frames

The known family of woven collections of frames have interesting properties and applications. Yet, they cannot be easily constructed. The present paper introduces new collections of frames sharing many properties with the original ones and having simpler construction. The frames {φi}i∈I and {ψi}i∈I for a Hilbert space H are called partition-woven (or briefly P-woven), if there exists a nontrivial subset σ of I such that the family {gi}i∈I represented as the triple ({φi},{ψi},σ) and called a P-weaven frame, is a frame defined by gi=φi if i∈σ and gi=ψi, otherwise. The paper also studies the class of CP-woven collections of frames which is a subset of the P-woven collections of frames and satisfy a stronger condition. The set of all woven frames is a strict subset of the set of all CP-woven frames which is further a strict subset of the set of all P-woven frames. Several examples and counterexamples are studied to show whether or not a pair of frames is P-woven or CP-woven.

Parametric study of seismic pounding between RC buildings with aligned slabs

This paper presents a parametrical study on seismic pounding between pairs of RC frames with 3 and 5 stories and aligned slabs. The performance indices are the maximum inter-story drift, absolute acceleration, and story shear force, and the cumulated hysteretic energy. The distinctive features are the selection of parameters and of performance indices, the soil-structure interaction, a modified Kelvin-Voigt pounding model, and, the careful selection of the discretization period. The findings of this study are the corroboration of remarks from previous researches and the formulation of new conclusions.

From Hessian to Weitzenböck: manifolds with torsion-carrying connections

We investigate affine connections that have zero curvature but not necessarily zero torsion. Slightly generalizing from what is known as Weitzenbock connections, such non-flat connections (we call “pseudo-Weitzenbock connections”) can be constructed from any frame (a set of linearly independent vector fields), not just the orthonormal frames, with their torsions vanishing if and only if the frame is a coordinate frame. In such situations, the notion of biorthogonal frames generalizes the notion of biorthogonal coordinates of a Hessian manifold, with respect to any given Riemannian metric g (not necessarily Hessian). Our main theorem shows that the pair of pseudo-Weitzenbock connections, each adapted to one of the pair of g-biorthogonal frames, are g-conjugate to each other. As a result, the pseudo-Weitzenbock connection pair generalize dually flat connections characteristic of Hessian manifolds, by being both curvature-free yet admitting (generally unequal) torsions. These results allow us to construct a pseudo-Weitzenbock connection for the manifold of parametric statistical models and treat it as a “statistical manifold admitting torsion”.

Study of Subjective Quality and Objective Blind Quality Prediction of Stereoscopic Videos.

We present a new subjective and objective study on full high-definition (HD) stereoscopic (3D or S3D) video quality. In subjective study, we constructed an S3D video dataset with 12 pristine and 288 test videos, and the test videos are generated by applying the H.264 and H.265 compression, blur and frame freeze artifacts. We also propose a no reference (NR) objective video quality assessment (QA) algorithm that relies on measurements of the statistical dependencies between the motion and disparity subband coefficients of S3D videos. Inspired by the Generalized Gaussian Distribution (GGD) approach in liu2011statistical, we model the joint statistical dependencies between the motion and disparity components as following a Bivariate Generalized Gaussian Distribution (BGGD). We estimate the BGGD model parameters (α,β) and the coherence measure (Ψ) from the eigenvalues of the sample covariance matrix (M) of the BGGD. In turn, we model the BGGD parameters of pristine S3D videos using a Multivariate Gaussian (MVG) distribution. The likelihood of a test video's MVG model parameters coming from the pristine MVG model is computed and shown to play a key role in the overall quality estimation. We also estimate the global motion content of each video by averaging the SSIM scores between pairs of successive video frames. To estimate the test S3D video's spatial quality, we apply the popular 2D NR unsupervised NIQE image QA model on a frame-by-frame basis on both views. The overall quality of a test S3D video is finally computed by pooling the test S3D video's likelihood estimates, global motion strength and spatial quality scores. The proposed algorithm, which is 'completely blind' (requiring no reference videos or training on subjective scores) is called the Motion and Disparity based 3D video quality evaluator (MoDi3D). We show that MoDi3D delivers competitive performance over a wide variety of datasets including the IRCCYN dataset, the WaterlooIVC Phase I dataset, the LFOVIA dataset and our proposed LFOVIAS3DPh2 S3D video dataset.

MOVING OBJECT DETECTION USING SPATIAL CORRELATION IN LAB COLOUR SPACE

Abstract. Background subtraction-based techniques of moving object detection are very common in computer vision programs. Each technique of background subtraction employs image thresholding algorithms. Different thresholding methods generate varying threshold values that provide dissimilar moving object detection results. A majority of background subtraction techniques use grey images which reduce the computational cost but statistics-based image thresholding methods do not consider the spatial distribution of pixels. In this study, authors have developed a background subtraction technique using Lab colour space and used spatial correlations for image thresholding. Four thresholding methods using spatial correlation are developed by computing the difference between opposite colour pairs of background and foreground frames. Out of 9 indoor and outdoor scenes, the object is detected successfully in 7 scenes whereas existing background subtraction technique using grey images with commonly used thresholding methods detected moving objects in 1–5 scenes. Shape and boundaries of detected objects are also better defined using the developed technique.

Learning-Based Sphere Nonlinear Interpolation for Motion Synthesis

Motion synthesis technology can produce natural and coordinated motion data without a motion capture process, which is complex and costly. Current motion synthesis methods usually provide a few interfaces to avoid the arbitrariness of the synthesis process, but this actually reduces the understandability of the synthesis process. In this paper, we propose a learning-based Sphere nonlinear interpolation (Snerp) model that can generate natural in-between motions in terms of a given start–end frame pair. Variety of the input frame pairs will enrich the diversity of the generated motions. The angle speed of natural human motion is not uniform and presents different change rules (we call them motion patterns) for different motions, so we first extract the motion patterns and then build the relation between motion pattern space and frame pair space via a paired dictionary learning process. After learning, we estimate the motion pattern according to the representation of a given start–end frame pair on the frame pair dictionary. We select several different types of start–end frame pairs from the real motion sequences as the testing data and good results of both objective and subjective evaluations on the generated motions demonstrate the superior performance of Snerp.

On-the-fly extrinsic calibration of multimodal sensing system for fast 3D thermographic scanning.

The fusion of three-dimensional (3D) geometrical and two-dimensional (2D) thermal information provides a promising method for characterizing temperature distribution of 3D objects, extending infrared imaging from 2D to 3D to support various thermal inspection applications. In this paper, we present an effective on-the-fly calibration approach for accurate alignment of depth and thermal data to facilitate dynamic and fast-speed 3D thermal scanning tasks. For each pair of depth and thermal frames, we estimate their relative pose by minimizing the objective function that measures the temperature consistency between a 2D infrared image and the reference 3D thermographic model. Our proposed frame-to-model mapping scheme can be seamlessly integrated into a generic 3D thermographic reconstruction framework. Through graphics-processing-unit-based acceleration, our method requires less than 10ms to generate a pair of well-aligned depth and thermal images without hardware synchronization and improves the robustness of the system against significant camera motion.

Joint Similarity and Parameterizations of Dilations of Dual $$(\Omega ,\mu )$$ ( Ω , μ ) -Frame Pairs

In this paper, we study some properties on the dilations of dual $$(\Omega ,\mu )$$ -frame pairs. Firstly, we introduce the concepts of joint complementary $$(\Omega ,\mu )$$ -frames and canonical dual $$(\Omega ,\mu )$$ -frame dilations of a dual $$(\Omega ,\mu )$$ -frame pair associated with an analysis space. Then we prove that the pairs of joint complementary $$(\Omega ,\mu )$$ -frames of a given dual $$(\Omega ,\mu )$$ -frame pair associated with an analysis space are unique in the sense of joint similarity and the set of canonical dual $$(\Omega ,\mu )$$ -frame dilations of a given dual $$(\Omega ,\mu )$$ -frame pair associated with an analysis space are parameterized by a set of invertible diagonal operators. Finally we parameterize the set of canonical dual $$(\Omega ,\mu )$$ -frame dilations of all dual $$(\Omega ,\mu )$$ -frame pairs of a given $$(\Omega ,\mu )$$ -frame associated with an analysis space in terms of a set of invertible lower triangular operators.

Tensor product dual frames

To construct dual frames with good structure for a given frame is a fundamental problem in the theory of frames. The tensor product duals of tensor product frames can provide a rank-one decomposition of bounded antilinear operators between two Hilbert spaces. This paper addresses tensor product dual frames. We derive a necessary and sufficient condition for two tensor product Bessel sequences to be a pair of dual frames; obtain explicit expressions of all dual frames and tensor product dual frames of tensor product frames; and demonstrate the existence of non-tensor product dual frames of tensor product frames.

Equiangular frames and generalizations of the Welch bound to dual pairs of frames

The purpose of this paper is twofold. First, we determine the lower bound for the maximum coherence between a pair of dual frames in and state conditions under which the lower bound is attained. It is shown that the existence of frames and duals that attain the lower bound is related to the existence of equiangular tight frames (ETFs). Second, motivated by the scarcity of ETFs (which by default have dual ETFs), we examine the more general question of existence of equiangular frames that have equiangular duals. For the case where an equiangular dual cannot be found we provide conditions under which the number of angles among vectors in the canonical dual frame is small.

Deep convolutional BiLSTM fusion network for facial expression recognition

Deep learning algorithms have shown significant performance improvements for facial expression recognition (FER). Most deep learning-based methods, however, focus more attention on spatial appearance features for classification, discarding much useful temporal information. In this work, we present a novel framework that jointly learns spatial features and temporal dynamics for FER. Given the image sequence of an expression, spatial features are extracted from each frame using a deep network, while the temporal dynamics are modeled by a convolutional network, which takes a pair of consecutive frames as input. Finally, the framework accumulates clues from fused features by a BiLSTM network. In addition, the framework is end-to-end learnable, and thus temporal information can be adapted to complement spatial features. Experimental results on three benchmark databases, CK+, Oulu-CASIA and MMI, show that the proposed framework outperforms state-of-the-art methods.

A method for dynamic segmentation of a pair of sequental video-frames

An algorithm of dynamic segmentation of sequential frame pairs was proposed. A comparative analysis of segmentation quality when finding shifts and affine inter-frame transformations for the segments was conducted. In addition, we compared the performance of the proposed method with some static segmentation approaches.