Discontinuity preserving in TV-L1 optical flow methods

[EN] We analyze the discontinuity preserving problem in TV-L1 optical flow methods. This type of methods typically creates rounded effects at flow boundaries, which usually do not coincide with object contours. A simple strategy to overcome this problem consists in inhibiting the diffusion at high image gradients. In this work, we first introduce a general framework for TV regularizers in optical flow and relate it with some standard approaches. Our survey takes into account several methods that use decreasing functions for mitigating the diffusion at image contours. Consequently, this kind of strategies may produce instabilities in the estimation of the optical flows. Hence, we study the problem of instabilities and show that it actually arises from an ill-posed formulation. From this study, it is possible to come across with different schemes to solve this problem. One of these consists in separating the pure TV process from the mitigating strategy. This has been used in another work and we demonstrate here that it has a good performance. Furthermore, we propose two alternatives to avoid the instability problems: (i) we study a fully automatic approach that solves the problem based on the information of the whole image; (ii) we derive a semi-automatic approach that takes into account the image gradients in a close neighborhood adapting the parameter in each position. In the experimental results, we present a detailed study and comparison between the different alternatives. These methods provide very good results, especially for sequences with a few dominant gradients. Additionally, a surprising effect of these approaches is that they can cope with occlusions. This can be easily achieved by using strong regularizations and high penalizations at image contours.

Regularization strategies for discontinuity-preserving optical flow methods

[EN]The aim of this work is to study several strategies for the preservation of flow discontinuities in variational optical flow methods. We analyze the combination of robust functionals and diffusion tensors in the smoothness assumption. Our study includes the use of tensors based on decreasing functions, which has shown to provide good results. However, it presents several limitations and usually does not perform better than other basic approaches. It typically introduces instabilities in the computed motion fields in the form of independent \textit{blobs} of vectors with large magnitude...

On the optical flow model selection through metaheuristics

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Evolutionary optimization applied for fine-tuning parameter estimation in optical flow-based environments

Optical flow methods are accurate algorithms for estimating the displacement and velocity fields of objects in a wide variety of applications, being their performance dependent on the configuration of a set of parameters. Since there is a lack of research that aims to automatically tune such parameters, in this work we have proposed an evolutionary-based framework for such task, thus introducing three techniques for such purpose: Particle Swarm Optimization, Harmony Search and Social-Spider Optimization. The proposed framework has been compared against with the well-known Large Displacement Optical Flow approach, obtaining the best results in three out eight image sequences provided by a public dataset. Additionally, the proposed framework can be used with any other optimization technique.

Parallel implementation of a robust optical flow technique

[EN] The accuracy and performance of current variational optical ow methods have considerably increased during the last years. The complexity of these techniques is high and enough care has to be taken for the implementation. The aim of this work is to present a comprehensible implementation of recent variational optical flow methods. We start with an energy model that relies on brightness and gradient constancy terms and a ow-based smoothness term. We minimize this energy model and derive an e cient implicit numerical scheme. In the experimental results, we evaluate the accuracy and performance of this implementation with the Middlebury benchmark database. We show that it is a competitive solution with respect to current methods in the literature. In order to increase the performance, we use a simple strategy to parallelize the execution on multi-core processors.

Horn-schunck optical flow with a multi-scale strategy

[EN] The seminal work of Horn and Schunck [8] is the first variational method for optical flow estimation. It introduced a novel framework where the optical flow is computed as the solution of a minimization problem. From the assumption that pixel intensities do not change over time, the optical flow constraint equation is derived. This equation relates the optical flow with the derivatives of the image. There are infinitely many vector fields that satisfy the optical flow constraint, thus the problem is ill-posed. To overcome this problem, Horn and Schunck introduced an additional regularity condition that restricts the possible solutions. Their method minimizes both the optical flow constraint and the magnitude of the variations of the flow field, producing smooth vector fields. One of the limitations of this method is that, typically, it can only estimate small motions. In the presence of large displacements, this method fails when the gradient of the image is not smooth enough. In this work, we describe an implementation of the original Horn and Schunck method and also introduce a multi-scale strategy in order to deal with larger displacements. For this multi-scale strategy, we create a pyramidal structure of downsampled images and change the optical flow constraint equation with a nonlinear formulation. In order to tackle this nonlinear formula, we linearize it and solve the method iteratively in each scale. In this sense, there are two common approaches: one that computes the motion increment in the iterations, like in ; or the one we follow, that computes the full flow during the iterations, like in. The solutions are incrementally refined ower the scales. This pyramidal structure is a standard tool in many optical flow methods.

Reliable estimation of dense optical flow fields with large displacements

[EN] In this paper we show that a classic optical flow technique by Nagel and Enkelmann can be regarded as an early anisotropic diffusion method with a diffusion tensor. We introduce three improvements into the model formulation that avoid inconsistencies caused by centering the brightness term and the smoothness term in different images use a linear scale-space focusing strategy from coarse to fine scales for avoiding convergence to physically irrelevant local minima, and create an energy functional that is invariant under linear brightness changes.  Applying a gradient descent method to the resulting energy functional leads to a system of diffusion-reaction equations. We prove that this system has a unique solution under realistic assumptions on the initial data, and we present an efficient linear implicit numerical scheme in detail. Our method creates flow fields with 100% density over the entire image domain, it is robust under a large range of parameter variations, and it can recover displacement fields that are far beyond the typical one-pixel limits which are characteristic for many differential methods for determining optical flow. We show that it performs better than the classic optical flow methods with 100%  density that are evaluated by Barron et al. (1994). Our software is available from the Internet.

Implementation of a robust optical flow method for color images

We analyse the influence of colour information in optical flow methods. Typically, most of these techniques compute their solutions using grayscale intensities due to its simplicity and faster processing, ignoring the colour features. However, the current processing systems have minimized their computational cost and, on the other hand, it is reasonable to assume that a colour image offers more details from the scene which should facilitate finding better flow fields. The aim of this work is to determine if a multi-channel approach supposes a quite enough improvement to justify its use. In order to address this evaluation, we use a multi-channel implementation of a well-known TV-L1 method. Furthermore, we review the state-of-the-art in colour optical flow methods. In the experiments, we study various solutions using grayscale and RGB images from recent evaluation datasets to verify the colour benefits in motion estimation.

Line search multilevel optimization as computational methods for dense optical flow

We evaluate the performance of different optimization techniques developed in the context of optical flowcomputation with different variational models. In particular, based on truncated Newton methods (TN) that have been an effective approach for large-scale unconstrained optimization, we develop the use of efficient multilevel schemes for computing the optical flow. More precisely, we evaluate the performance of a standard unidirectional multilevel algorithm - called multiresolution optimization (MR/OPT), to a bidrectional multilevel algorithm - called full multigrid optimization (FMG/OPT). The FMG/OPT algorithm treats the coarse grid correction as an optimization search direction and eventually scales it using a line search. Experimental results on different image sequences using four models of optical flow computation show that the FMG/OPT algorithm outperforms both the TN and MR/OPT algorithms in terms of the computational work and the quality of the optical flow estimation.

Line search multilevel optimization as computational methods for dense optical flow

We evaluate the performance of different optimization techniques developed in the context of optical flow computation with different variational models. In particular, based on truncated Newton methods (TN) that have been an effective approach for large-scale unconstrained optimization, we de- velop the use of efficient multilevel schemes for computing the optical flow. More precisely, we evaluate the performance of a standard unidirectional mul- tilevel algorithm - called multiresolution optimization (MR/OPT), to a bidrec- tional multilevel algorithm - called full multigrid optimization (FMG/OPT). The FMG/OPT algorithm treats the coarse grid correction as an optimiza- tion search direction and eventually scales it using a line search. Experimental results on different image sequences using four models of optical flow com- putation show that the FMG/OPT algorithm outperforms both the TN and MR/OPT algorithms in terms of the computational work and the quality of the optical flow estimation.

Computing inverse optical flow

[EN] We propose four algorithms for computing the inverse optical flow between two images. We assume that the forward optical flow has already been obtained and we need to estimate the flow in the backward direction. The forward and backward flows can be related through a warping formula, which allows us to propose very efficient algorithms. These are presented in increasing order of complexity. The proposed methods provide high accuracy with low memory requirements and low running times.In general, the processing reduces to one or two image passes. Typically, when objects move in a sequence, some regions may appear or disappear. Finding the inverse flows in these situations is difficult and, in some cases, it is not possible to obtain a correct solution. Our algorithms deal with occlusions very easy and reliably. On the other hand, disocclusions have to be overcome as a post-processing step. We propose three approaches for filling disocclusions. In the experimental results, we use standard synthetic sequences to study the performance of the proposed methods, and show that they yield very accurate solutions. We also analyze the performance of the filling strategies. 

A scale-space aproach to nonlocal optical flow calculations

[EN] This paper presents an interpretation of a classic optical flow method by Nagel and Enkelmann as a tensor-driven anisotropic diffusion approach in digital image analysis. We introduce an improvement into the model formulation, and we establish well-posedness results for the resulting system of parabolic partial differential equations. Our method avoids linearizations in the optical flow constraint, and it can recover displacement fields which are far beyond the typical one-pixel limits that are characteristic for many differential methods for optical flow recovery. A robust numerical scheme is presented in detail. We avoid convergence to irrelevant local minima by embedding our method into a linear scale-space framework and using a focusing strategy from coarse to fine scales. The high accuracy of the proposed method is demonstrated by means of a synthetic and a real-world image sequence.

A PDE model for computing the optical flow

[EN] In this paper we present a new model for optical flow calculation using a variational formulation which preserves discontinuities of the flow much better than classical methods. We study the Euler-Lagrange equations asociated to the variational problem. In the case of quadratic energy, we show the existence and uniqueness of the corresponding evolution problem. Since our method avoid linearization in the optical flow constraint, it can recover large displacement in the scene. We avoid convergence to irrelevant local minima by embedding our method into a linear scale-space framework and using a focusing strategy from coarse to fine scales.

Evaluation of the carbon content of aerosols from the burning of biomass in the Brazilian Amazon using thermal, optical and thermal-optical analysis methods

Aerosol samples were collected at a pasture site in the Amazon Basin as part of the project LBA-SMOCC-2002 (Large-Scale Biosphere-Atmosphere Experiment in Amazonia - Smoke Aerosols, Clouds, Rainfall and Climate: Aerosols from Biomass Burning Perturb Global and Regional Climate). Sampling was conducted during the late dry season, when the aerosol composition was dominated by biomass burning emissions, especially in the submicron fraction. A 13-stage Dekati low-pressure impactor (DLPI) was used to collect particles with nominal aerodynamic diameters (D(p)) ranging from 0.03 to 0.10 mu m. Gravimetric analyses of the DLPI substrates and filters were performed to obtain aerosol mass concentrations. The concentrations of total, apparent elemental, and organic carbon (TC, EC(a), and OC) were determined using thermal and thermal-optical analysis (TOA) methods. A light transmission method (LTM) was used to determine the concentration of equivalent black carbon (BC(e)) or the absorbing fraction at 880 nm for the size-resolved samples. During the dry period, due to the pervasive presence of fires in the region upwind of the sampling site, concentrations of fine aerosols (D(p) < 2.5 mu m: average 59.8 mu g m(-3)) were higher than coarse aerosols (D(p) > 2.5 mu m: 4.1 mu g m(-3)). Carbonaceous matter, estimated as the sum of the particulate organic matter (i.e., OC x 1.8) plus BC(e), comprised more than 90% to the total aerosol mass. Concentrations of EC(a) (estimated by thermal analysis with a correction for charring) and BC(e) (estimated by LTM) averaged 5.2 +/- 1.3 and 3.1 +/- 0.8 mu g m(-3), respectively. The determination of EC was improved by extracting water-soluble organic material from the samples, which reduced the average light absorption Angstrom exponent of particles in the size range of 0.1 to 1.0 mu m from >2.0 to approximately 1.2. The size-resolved BC(e) measured by the LTM showed a clear maximum between 0.4 and 0.6 mu m in diameter. The concentrations of OC and BC(e) varied diurnally during the dry period, and this variation is related to diurnal changes in boundary layer thickness and in fire frequency.

An optical flow-based sensing system for reactive mobile robot navigation

This work discusses the use of optical flow to generate the sensorial information a mobile robot needs to react to the presence of obstacles when navigating in a non-structured environment. A sensing system based on optical flow and time-to-collision calculation is here proposed and experimented, which accomplishes two important paradigms. The first one is that all computations are performed onboard the robot, in spite of the limited computational capability available. The second one is that the algorithms for optical flow and time-to-collision calculations are fast enough to give the mobile robot the capability of reacting to any environmental change in real-time. Results of real experiments in which the sensing system here proposed is used as the only source of sensorial data to guide a mobile robot to avoid obstacles while wandering around are presented, and the analysis of such results allows validating the proposed sensing system.