Momo: Monocular motion estimation on manifolds

Graeter, Johannes; Strauß, Tobias; Lauer, Martin

doi:10.1109/itsc.2017.8317679

Cited by 8 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [14], a visual odometry method was developed to calculate the movement of a monocular camera with six degrees of freedom, assuming that the ground plane is known. In [15], a robust new visual odometry framework was presented by considering the motion model of the vehicle. In [16], an end-to-end deep learning framework was used to train a regressor for visual odometry.…”

Section: Related Workmentioning

confidence: 99%

Exploring Self-Attention for Visual Intersection Classification

Nakata¹,

Takeda²

2022

Preprint

View full text Add to dashboard Cite

In robot vision, self-attention has recently emerged as a technique for capturing non-local contexts. In this study, we introduced a self-attention mechanism into the intersection recognition system as a method to capture the non-local contexts behind the scenes. An intersection classification system comprises two distinctive modules: (a) a first-person vision (FPV) module, which uses a short egocentric view sequence as the intersection is passed, and (b) a third-person vision (TPV) module, which uses a single view immediately before entering the intersection. The self-attention mechanism is effective in the TPV module because most parts of the local pattern (e.g., road edges, buildings, and sky) are similar to each other, and thus the use of a non-local context (e.g., the angle between two diagonal corners around an intersection) would be effective. This study makes three major contributions. First, we proposed a self-attention-based approach for intersection classification using TPVs. Second, we presented a practical system in which a self-attention-based TPV module is combined with an FPV module to improve the overall recognition performance. Finally, experiments using the public KITTI dataset show that the above self-attention-based system outperforms conventional recognition based on local patterns and recognition based on convolution operations.

show abstract

Section: Related Workmentioning

confidence: 99%

Exploring Self-Attention for Visual Intersection Classification

Nakata¹,

Takeda²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…We use semantics and cheirality for a preliminary outlier rejection as mentioned in Section V-C. Though most of the outliers can be detected by these methods, some still remain such as moving shadows of vehicles, non classified moving objects, etc.. As shown in our previous work [25], loss functions improve estimation accuracy and robustness drastically. Therefore, we employ the Cauchy function as loss functions ρ φ (x), ρ ξ (x) in order to reduce the influence of large residuals in both, the depth and the reprojection error cost functions.…”

Section: E Robustification and Problem Formulationmentioning

confidence: 99%

LIMO: Lidar-Monocular Visual Odometry

Graeter

Wilczynski

Lauer

2018

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Self Cite

196

View full text Add to dashboard Cite

Higher level functionality in autonomous driving depends strongly on a precise motion estimate of the vehicle. Powerful algorithms have been developed. However, their great majority focuses on either binocular imagery or pure LIDAR measurements. The promising combination of camera and LI-DAR for visual localization has mostly been unattended. In this work we fill this gap, by proposing a depth extraction algorithm from LIDAR measurements for camera feature tracks and estimating motion by robustified keyframe based Bundle Adjustment. Semantic labeling is used for outlier rejection and weighting of vegetation landmarks. The capability of this sensor combination is demonstrated on the competitive KITTI dataset, achieving a placement among the top 15. The code is released to the community.

show abstract

“…b) Based on Epipolar Geometry: Having obtained a set of 2D correspondences of the scene, the ego motion can be estimated up to scale. Therefore, the motion estimate lies on a manifold with five degrees of freedom [12] which are the three angles of rotation and the two translation vector directions. Classically, an error metric that is based on the epipolar geometry is used for optimization (see [13], [14]).…”

Section: B Odometry Estimationmentioning

confidence: 99%

“…c) Transformation Estimation: To reliably estimate a transformation between two frames, we use the weighted least squares method presented by Marx [11] which was introduced in Section II-B. This method yields an orthonormal rotation matrix R2←1 and a translation t2←1 so that we can estimate the motion flow fmotion,2←1 (x 1 ) = R2←1 x 1 + t2←1 (12) as defined in Section III-B. Based on this scene flow, a rigid-body transformation is estimated in the training process that we transform into a motion flow field (see Fig.…”

Section: A Implementationmentioning

confidence: 99%

Self-Supervised Flow Estimation using Geometric Regularization with Applications to Camera Image and Grid Map Sequences

Wirges

Gräter

Zhang

et al. 2019

2019 IEEE Intelligent Transportation Systems Conference (ITSC)

View full text Add to dashboard Cite

We present a self-supervised approach to estimate flow in camera image and top-view grid map sequences using fully convolutional neural networks in the domain of automated driving. We extend existing approaches for self-supervised optical flow estimation by adding a regularizer expressing motion consistency assuming a static environment. However, as this assumption is violated for other moving traffic participants we also estimate a mask to scale this regularization. Adding a regularization towards motion consistency improves convergence and flow estimation accuracy. Furthermore, we scale the errors due to spatial flow inconsistency by a mask that we derive from the motion mask. This improves accuracy in regions where the flow drastically changes due to a better separation between static and dynamic environment. We apply our approach to optical flow estimation from camera image sequences, validate on odometry estimation and suggest a method to iteratively increase optical flow estimation accuracy using the generated motion masks. Finally, we provide quantitative and qualitative results based on the KITTI odometry and tracking benchmark for scene flow estimation based on grid map sequences. We show that we can improve accuracy and convergence when applying motion and spatial consistency regularization.

show abstract

Momo: Monocular motion estimation on manifolds

Cited by 8 publications

References 17 publications

Exploring Self-Attention for Visual Intersection Classification

Exploring Self-Attention for Visual Intersection Classification

LIMO: Lidar-Monocular Visual Odometry

Self-Supervised Flow Estimation using Geometric Regularization with Applications to Camera Image and Grid Map Sequences

Contact Info

Product

Resources

About