Motion estimation is one of the major problems in developing video coding applications. Among all motion estimation approaches, Block-matching (BM) algorithms are the most popular methods due to their effectiveness and simplicity for both software and hardware implementations. A BM approach assumes that the movement of pixels within a defined region of the current frame can be modeled as a translation of pixels contained in the previous frame. In this procedure, the motion vector is obtained by minimizing a certain matching metric that is produced for the current frame over a determined search window from the previous frame. Unfortunately, the evaluation of such matching measurement is computationally expensive and represents the most consuming operation in the BM process. Therefore, BM motion estimation can be viewed as an optimization problem whose goal is to find the best-matching block within a search space. The simplest available BM method is the Full Search Algorithm (FSA) which finds the most accurate motion vector through an exhaustive computation of all the elements of the search space. Recently, several fast BM algorithms have been proposed to reduce the search positions by calculating only a fixed subset of motion vectors despite lowering its accuracy. On the other hand, the Harmony Search (HS) algorithm is a population-based optimization method that is inspired by the music improvisation process in which a musician searches for harmony and continues to polish the pitches to obtain a better harmony. In this paper, a new BM algorithm that combines HS with a fitness approximation model is proposed. The approach uses motion vectors belonging to the search window as potential solutions. A fitness function evaluates the matching quality of each motion vector candidate.
We extend the concept of optical flow with spatiotemporal regularisation to a dynamic non-Euclidean setting. Optical flow is traditionally computed from a sequence of flat images. The purpose of this paper is to introduce variational motion estimation for images that are defined on an evolving surface. Volumetric microscopy images depicting a live zebrafish embryo serve as both biological motivation and test data.
Block matching (BM) motion estimation plays a very important role in video coding. In a BM approach, image frames in a video sequence are divided into blocks. For each block in the current frame, the best matching block is identified inside a region of the previous frame, aiming to minimize the sum of absolute differences (SAD). Unfortunately, the SAD evaluation is computationally expensive and represents the most consuming operation in the BM process. Therefore, BM motion estimation can be approached as an optimization problem, where the goal is to find the best matching block within a search space. The simplest available BM method is the full search algorithm (FSA) which finds the most accurate motion vector through an exhaustive computation of SAD values for all elements of the search window. Recently, several fast BM algorithms have been proposed to reduce the number of SAD operations by calculating only a fixed subset of search locations at the price of poor accuracy. In this paper, a new algorithm based on Artificial Bee Colony (ABC) optimization is proposed to reduce the number of search locations in the BM process. In our algorithm, the computation of search locations is drastically reduced by considering a fitness calculation strategy which indicates when it is feasible to calculate or only estimate new search locations. Since the proposed algorithm does not consider any fixed search pattern or any other movement assumption as most of other BM approaches do, a high probability for finding the true minimum (accurate motion vector) is expected. Conducted simulations show that the proposed method achieves the best balance over other fast BM algorithms, in terms of both estimation accuracy and computational cost.
In this paper we present a number of methods (manual, semi-automatic and automatic) for tracking individual targets in high density crowd scenes where thousand of people are gathered. The necessary data about the motion of individuals and a lot of other physical information can be extracted from consecutive image sequences in different ways, including optical flow and block motion estimation. One of the famous methods for tracking moving objects is the block matching method. This way to estimate subject motion requires the specification of a comparison window which determines the scale of the estimate. In this work we present a real-time method for pedestrian recognition and tracking in sequences of high resolution images obtained by a stationary (high definition) camera located in different places on the Haram mosque in Mecca. The objective is to estimate pedestrian velocities as a function of the local density.The resulting data of tracking moving pedestrians based on video sequences are presented in the following section. Through the evaluated system the spatio-temporal coordinates of each pedestrian during the Tawaf ritual are established. The pilgrim velocities as function of the local densities in the Mataf area (Haram Mosque Mecca) are illustrated and very precisely documented.
In object tracking, outlier is one of primary factors which degrade performance of image-based tracking algorithms. In this respect, therefore, most of the existing methods simply discard detected outliers and pay little or no attention to employing them as an important source of information for motion estimation. We consider outliers as important as inliers for object tracking and propose a motion estimation algorithm based on concurrent tracking of inliers and outliers. Our tracker makes use of pyramidal implementation of the Lucas-Kanade tracker to estimate motion flows of inliers and outliers and final target motion is estimated robustly based on both of these information. Experimental results from challenging benchmark video sequences confirm enhanced tracking performance, showing highly stable target tracking under severe occlusion compared with state-of-the-art algorithms. The proposed algorithm runs at more than 100 frames per second even without using a hardware accelerator, which makes the proposed method more practical and portable.
In this work, we address the problem of outlier detection for robust motion estimation by using modern sparse-low-rank decompositions, i.e., Robust PCA-like methods, to impose global rank constraints. Robust decompositions have shown to be good at splitting a corrupted matrix into an uncorrupted low-rank matrix and a sparse matrix, containing outliers. However, this process only works when matrices have relatively low rank with respect to their ambient space, a property not met in motion estimation problems. As a solution, we propose to exploit the partial information present in the decomposition to decide which matches are outliers. We provide evidences showing that even when it is not possible to recover an uncorrupted low-rank matrix, the resulting information can be exploited for outlier detection. To this end we propose the Robust Decomposition with Constrained Rank (RD-CR), a proximal gradient based method that enforces the rank constraints inherent to motion estimation. We also present a general framework to perform robust estimation for stereo Visual Odometry, based on our RD-CR and a simple but effective compressed optimization method that achieves high performance. Our evaluation on synthetic data and on the KITTI dataset demonstrates the applicability of our approach in complex scenarios and it yields state-of-the-art performance.
Point sets matching method is very important in computer vision, feature extraction, fingerprint matching, motion estimation and so on. This paper proposes a robust point sets matching method. We present an iterative algorithm that is robust to noise case. Firstly, we calculate all transformations between two points. Then similarity matrix are computed to measure the possibility that two transformation are both true. We iteratively update the matching score matrix by using the similarity matrix. By using matching algorithm on graph, we obtain the matching result. Experimental results obtained by our approach show robustness to outlier and jitter.
We study the inverse problem of estimating n locations $t_1, ..., t_n$ (up to global scale, translation and negation) in $R^d$ from noisy measurements of a subset of the (unsigned) pairwise lines that connect them, that is, from noisy measurements of $\pm (t_i - t_j)/\|t_i - t_j\|$ for some pairs (i,j) (where the signs are unknown). This problem is at the core of the structure from motion (SfM) problem in computer vision, where the $t_i$'s represent camera locations in $R^3$. The noiseless version of the problem, with exact line measurements, has been considered previously under the general title of parallel rigidity theory, mainly in order to characterize the conditions for unique realization of locations. For noisy pairwise line measurements, current methods tend to produce spurious solutions that are clustered around a few locations. This sensitivity of the location estimates is a well-known problem in SfM, especially for large, irregular collections of images. In this paper we introduce a semidefinite programming (SDP) formulation, specially tailored to overcome the clustering phenomenon. We further identify the implications of parallel rigidity theory for the location estimation problem to be well-posed, and prove exact (in the noiseless case) and stable location recovery results. We also formulate an alternating direction method to solve the resulting semidefinite program, and provide a distributed version of our formulation for large numbers of locations. Specifically for the camera location estimation problem, we formulate a pairwise line estimation method based on robust camera orientation and subspace estimation. Lastly, we demonstrate the utility of our algorithm through experiments on real images.
In this paper, a macroblock classification method is proposed for various video processing applications involving motions. Based on the analysis of the Motion Vector field in the compressed video, we propose to classify Macroblocks of each video frame into different classes and use this class information to describe the frame content. We demonstrate that this low-computation-complexity method can efficiently catch the characteristics of the frame. Based on the proposed macroblock classification, we further propose algorithms for different video processing applications, including shot change detection, motion discontinuity detection, and outlier rejection for global motion estimation. Experimental results demonstrate that the methods based on the proposed approach can work effectively on these applications.
This paper introduces a state-of-the-art video representation and applies it to efficient action recognition and detection. We first propose to improve the popular dense trajectory features by explicit camera motion estimation. More specifically, we extract feature point matches between frames using SURF descriptors and dense optical flow. The matches are used to estimate a homography with RANSAC. To improve the robustness of homography estimation, a human detector is employed to remove outlier matches from the human body as human motion is not constrained by the camera. Trajectories consistent with the homography are considered as due to camera motion, and thus removed. We also use the homography to cancel out camera motion from the optical flow. This results in significant improvement on motion-based HOF and MBH descriptors. We further explore the recent Fisher vector as an alternative feature encoding approach to the standard bag-of-words histogram, and consider different ways to include spatial layout information in these encodings. We present a large and varied set of evaluations, considering (i) classification of short basic actions on six datasets, (ii) localization of such actions in feature-length movies, and (iii) large-scale recognition of complex events. We find that our improved trajectory features significantly outperform previous dense trajectories, and that Fisher vectors are superior to bag-of-words encodings for video recognition tasks. In all three tasks, we show substantial improvements over the state-of-the-art results.
In a range of fields including the geosciences, molecular biology, robotics and computer vision, one encounters problems that involve random variables on manifolds. Currently, there is a lack of flexible probabilistic models on manifolds that are fast and easy to train. We define an extremely flexible class of exponential family distributions on manifolds such as the torus, sphere, and rotation groups, and show that for these distributions the gradient of the log-likelihood can be computed efficiently using a non-commutative generalization of the Fast Fourier Transform (FFT). We discuss applications to Bayesian camera motion estimation (where harmonic exponential families serve as conjugate priors), and modelling of the spatial distribution of earthquakes on the surface of the earth. Our experimental results show that harmonic densities yield a significantly higher likelihood than the best competing method, while being orders of magnitude faster to train.
This paper addresses the problem of compression of 3D point cloud sequences that are characterized by moving 3D positions and color attributes. As temporally successive point cloud frames are similar, motion estimation is key to effective compression of these sequences. It however remains a challenging problem as the point cloud frames have varying numbers of points without explicit correspondence information. We represent the time-varying geometry of these sequences with a set of graphs, and consider 3D positions and color attributes of the points clouds as signals on the vertices of the graphs. We then cast motion estimation as a feature matching problem between successive graphs. The motion is estimated on a sparse set of representative vertices using new spectral graph wavelet descriptors. A dense motion field is eventually interpolated by solving a graph-based regularization problem. The estimated motion is finally used for removing the temporal redundancy in the predictive coding of the 3D positions and the color characteristics of the point cloud sequences. Experimental results demonstrate that our method is able to accurately estimate the motion between consecutive frames. Moreover, motion estimation is shown to bring significant improvement in terms of the overall compression performance of the sequence. To the best of our knowledge, this is the first paper that exploits both the spatial correlation inside each frame (through the graph) and the temporal correlation between the frames (through the motion estimation) to compress the color and the geometry of 3D point cloud sequences in an efficient way.
Super-resolution without explicit sub-pixel motion estimation is a very active subject of image reconstruction containing general motion. The Non-Local Means (NLM) method is a simple image reconstruction method without explicit motion estimation. In this paper we generalize NLM method to higher orders using kernel regression can apply to super-resolution reconstruction. The performance of the generalized method is compared with other methods.
Visual motion estimation is a computationally intensive, but important task for sighted animals. Replicating the robustness and efficiency of biological visual motion estimation in artificial systems would significantly enhance the capabilities of future robotic agents. 25 years ago, in this very journal, Carver Mead outlined his argument for replicating biological processing in silicon circuits. His vision served as the foundation for the field of neuromorphic engineering, which has experienced a rapid growth in interest over recent years as the ideas and technologies mature. Replicating biological visual sensing was one of the first tasks attempted in the neuromorphic field. In this paper we focus specifically on the task of visual motion estimation. We describe the task itself, present the progression of works from the early first attempts through to the modern day state-of-the-art, and provide an outlook for future directions in the field.
Head movement during scanning impedes activation detection in fMRI studies. Head motion in fMRI acquired using slice-based Echo Planar Imaging (EPI) can be estimated and compensated by aligning the images onto a reference volume through image registration. However, registering EPI images volume to volume fails to consider head motion between slices, which may lead to severely biased head motion estimates. Slice-to-volume registration can be used to estimate motion parameters for each slice by more accurately representing the image acquisition sequence. However, accurate slice to volume mapping is dependent on the information content of the slices: middle slices are information rich, while edge slides are information poor and more prone to distortion. In this work, we propose a Gaussian particle filter based head motion tracking algorithm to reduce the image misregistration errors. The algorithm uses a dynamic state space model of head motion with an observation equation that models continuous slice acquisition of the scanner. Under this model the particle filter provides more accurate motion estimates and voxel position estimates. We demonstrate significant performance improvement of the proposed approach as compared to registration-only methods of head motion estimation and brain activation detection.
The aim of this paper is to discuss and evaluate total variation based regularization methods for motion estimation, with particular focus on optical flow models. In addition to standard $L^2$ and $L^1$ data fidelities we give an overview of different variants of total variation regularization obtained from combination with higher order models and a unified computational optimization approach based on primal-dual methods. Moreover, we extend the models by Bregman iterations and provide an inverse problems perspective to the analysis of variational optical flow models. A particular focus of the paper is the quantitative evaluation of motion estimation, which is a difficult and often underestimated task. We discuss several approaches for quality measures of motion estimation and apply them to compare the previously discussed regularization approaches.
A tracking system that will be used for Augmented Reality (AR) applications has two main requirements: accuracy and frame rate. The first requirement is related to the performance of the pose estimation algorithm and how accurately the tracking system can find the position and orientation of the user in the environment. Accuracy problems of current tracking devices, considering that they are low-cost devices, cause static errors during this motion estimation process. The second requirement is related to dynamic errors (the end-to-end system delay; occurring because of the delay in estimating the motion of the user and displaying images based on this estimate. This paper investigates combining the vision-based estimates with measurements from other sensors, GPS and IMU, in order to improve the tracking accuracy in outdoor environments. The idea of using Fuzzy Adaptive Multiple Models (FAMM) was investigated using a novel fuzzy rule-based approach to decide on the model that results in improved accuracy and faster convergence for the fusion filter. Results show that the developed tracking system is more accurate than a conventional GPS-IMU fusion approach due to additional estimates from a camera and fuzzy motion models. The paper also presents an application in cultural heritage context.
Traditional methods for motion estimation estimate the motion field F between a pair of images as the one that minimizes a predesigned cost function. In this paper, we propose a direct method and train a Convolutional Neural Network (CNN) that when, at test time, is given a pair of images as input it produces a dense motion field F at its output layer. In the absence of large datasets with ground truth motion that would allow classical supervised training, we propose to train the network in an unsupervised manner. The proposed cost function that is optimized during training, is based on the classical optical flow constraint. The latter is differentiable with respect to the motion field and, therefore, allows backpropagation of the error to previous layers of the network. Our method is tested on both synthetic and real image sequences and performs similarly to the state-of-the-art methods.
This paper shows how to extract dense optical flow from videos with a convolutional neural network (CNN). The proposed model constitutes a potential building block for deeper architectures to allow using motion without resorting to an external algorithm, \eg for recognition in videos. We derive our network architecture from signal processing principles to provide desired invariances to image contrast, phase and texture. We constrain weights within the network to enforce strict rotation invariance and substantially reduce the number of parameters to learn. We demonstrate end-to-end training on only 8 sequences of the Middlebury dataset, orders of magnitude less than competing CNN-based motion estimation methods, and obtain comparable performance to classical methods on the Middlebury benchmark. Importantly, our method outputs a distributed representation of motion that allows representing multiple, transparent motions, and dynamic textures. Our contributions on network design and rotation invariance offer insights nonspecific to motion estimation.
We propose robust methods for estimating camera egomotion in noisy, real-world monocular image sequences in the general case of unknown observer rotation and translation with two views and a small baseline. This is a difficult problem because of the nonconvex cost function of the perspective camera motion equation and because of non-Gaussian noise arising from noisy optical flow estimates and scene non-rigidity. To address this problem, we introduce the expected residual likelihood method (ERL), which estimates confidence weights for noisy optical flow data using likelihood distributions of the residuals of the flow field under a range of counterfactual model parameters. We show that ERL is effective at identifying outliers and recovering appropriate confidence weights in many settings. We compare ERL to a novel formulation of the perspective camera motion equation using a lifted kernel, a recently proposed optimization framework for joint parameter and confidence weight estimation with good empirical properties. We incorporate these strategies into a motion estimation pipeline that avoids falling into local minima. We find that ERL outperforms the lifted kernel method and baseline monocular egomotion estimation strategies on the challenging KITTI dataset, while adding almost no runtime cost over baseline egomotion methods.