The estimation of the optical flow between two images is one of the key problems in low-level vision. According the optical flow evaluation site at http://vision.middlebury.edu/flow/, discontinuity preserving variational models based on Total Variation (TV) regularization and L 1 data terms are among the most accurate flow estimation techniques, but there is still room for improvements.
Fig. 1. Optical flow for the backyard and mini cooper scene of the Middlebury optical flow benchmark. Optical flow captures the dynamics of a scene by estimating the motion of every pixel between two frames of an image sequence. The displacement of every pixel is shown as displacement vectors on top of the commonly used flow color scheme (see Figure 5).
Abstract.A look at the Middlebury optical flow benchmark  reveals that nowadays variational methods yield the most accurate optical flow fields between two image frames. In this work we propose an improvement variant of the original duality based TV-L 1 optical flow algorithm in  and provide implementation details. This formulation can preserve discontinuities in the flow field by employing total variation (TV) regularization. Furthermore, it offers robustness against outliers by applying the robust L 1 norm in the data fidelity term.Our contributions are as follows. First, we propose to perform a structure-texture decomposition of the input images to get rid of violations in the optical flow constraint due to illumination changes. Second, we propose to integrate a median filter into the numerical scheme to further increase the robustness to sampling artefacts in the image data. We experimentally show that very precise and robust estimation of optical flow can be achieved with a variational approach in realtime. The numerical scheme and the implementation are described in a detailed way, which enables reimplementation of this high-end method.
This paper presents a technique for estimating the threedimensional velocity vector field that describes the motion of each visible scene point (scene flow). The technique presented uses two consecutive image pairs from a stereo sequence. The main contribution is to decouple the position and velocity estimation steps, and to estimate dense velocities using a variational approach. We enforce the scene flow to yield consistent displacement vectors in the left and right images. The decoupling strategy has two main advantages: Firstly, we are independent in choosing a disparity estimation technique, which can yield either sparse or dense correspondences, and secondly, we can achieve frame rates of 5 fps on standard consumer hardware. The approach provides dense velocity estimates with accurate results at distances up to 50 meters.
Building upon recent developments in optical flow and stereo matching estimation, we propose a variational framework for the estimation of stereoscopic scene flow, i.e., the motion of points in the three-dimensional world from stereo image sequences. The proposed algorithm takes into account image pairs from two consecutive times and computes both depth and a 3D motion vector associated with each point in the image. In contrast to previous works, we partially decouple the depth estimation from the motion estimation, which has many practical advantages. The variational formulation is quite flexible and can handle both sparse or dense disparity maps. The proposed method is very efficient; with the depth map being computed on an FPGA, and the scene flow computed on the GPU, the proposed algorithm runs at frame rates of 20 frames per second on QVGA images (320 × 240 pixels). Furthermore, we present solutions to two important problems in scene flow estimation: violations of intensity consistency between input images, and the uncertainty measures for the scene flow result.A. Wedel ( ) · C. Rabe · U. Franke GR/PAA, Daimler Research,
Understanding traffic situations in dynamic traffic environments is an essential requirement for autonomous driving. The prediction of the current traffic scene into the future is one of the main problems in this context. In this publication we focus on highway scenarios, where the maneuver space for traffic participants is limited to a small number of possible behavior classes. Even though there are many publications in the field of maneuver prediction, most of them set the focus on the classification problem, whether a certain maneuver is executed or not. We extend approaches which solve the classification problem of lane-change behavior by introducing the novel aspect of estimating a continuous distribution of possible trajectories.Our novel approach uses the probabilities which are assigned by a Random Decision Forest to each of the maneuvers lane following, lane change left and lane change right. Using measured data of a vehicle and the knowledge of the typical lateral movement of vehicles over time taken from realworlddata, we derive a Gaussian Mixture Regression method. For the final result we combine the predicted probability density functions of the regression method and the computed maneuver probabilities using a Mixture of Experts approach.In a large scale experiment on real world data collected on multiple test drives we trained and validated our prediction model and show the gained high prediction accuracy of the proposed method.
scite is a Brooklyn-based startup that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.