Globally-Optimal Event Camera Motion Estimation

Peng, Xin; Wang, Yifu; Gao, Ling; Kneip, Laurent

doi:10.1007/978-3-030-58574-7_4

Cited by 22 publications

(13 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous studies have shown that contrast maximization and entropy minimization can cause the loss function to have multiple extremes in some situations [33,34]. In this study, we clearly showed that when using contrast maximization and entropy minimization to perform homographic motion estimation for a plane, the loss function has multiple extremes and the motion estimation result deviates from the true value.…”

Section: Discussionmentioning

confidence: 57%

“…In response to this problem, Liu et al [33] proposed a method for finding the global maximum solution of a rotational motion using a branch and bound (BnB) approach for contrast maximization. Peng et al [34] also proposed a method for finding the global maximum solution using BnB and achieved planar motion estimation from an event camera mounted downward on an automatic guided vehicle (AGV). However, when mounting an event camera on a car, it is difficult to capture the road surface vertically.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Accuracy and Speed Improvement of Event Camera Motion Estimation Using a Bird’s-Eye View Transformation

Ozawa

Sekikawa

Saitō

2022

Sensors

View full text Add to dashboard Cite

Event cameras are bio-inspired sensors that have a high dynamic range and temporal resolution. This property enables motion estimation from textures with repeating patterns, which is difficult to achieve with RGB cameras. Therefore, motion estimation of an event camera is expected to be applied to vehicle position estimation. An existing method, called contrast maximization, is one of the methods that can be used for event camera motion estimation by capturing road surfaces. However, contrast maximization tends to fall into a local solution when estimating three-dimensional motion, which makes correct estimation difficult. To solve this problem, we propose a method for motion estimation by optimizing contrast in the bird’s-eye view space. Instead of performing three-dimensional motion estimation, we reduced the dimensionality to two-dimensional motion estimation by transforming the event data to a bird’s-eye view using homography calculated from the event camera position. This transformation mitigates the problem of the loss function becoming non-convex, which occurs in conventional methods. As a quantitative experiment, we created event data by using a car simulator and evaluated our motion estimation method, showing an improvement in accuracy and speed. In addition, we conducted estimation from real event data and evaluated the results qualitatively, showing an improvement in accuracy.

show abstract

Section: Discussionmentioning

confidence: 57%

Section: Related Workmentioning

confidence: 99%

Accuracy and Speed Improvement of Event Camera Motion Estimation Using a Bird’s-Eye View Transformation

Ozawa

Sekikawa

Saitō

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Other works constrain the problem to car-like vehicles [11][12][13]. By assuming the Ackermann steering model, the general planar motion can be restricted to 2DoF.…”

Section: Related Workmentioning

confidence: 99%

“…Gao et al [11] propose a keypoint-less method to recover the ego motion of car-like steering models using an optimal image registration technique based on global branch-and-bound optimization. Peng et al [12] use an event-based camera system to overcome the downsides of a frame-based camera such as latency, motion blur and low dynamic range. In [13], template matching is applied for calculating the image displacement between consecutive frames.…”

Section: Related Workmentioning

confidence: 99%

Unsupervised deep learning based ego motion estimation with a downward facing camera

Gilles

Ibrahimpasic²

2021

Vis Comput

View full text Add to dashboard Cite

Knowing the robot's pose is a crucial prerequisite for mobile robot tasks such as collision avoidance or autonomous navigation. Using powerful predictive models to estimate transformations for visual odometry via downward facing cameras is an understudied area of research. This work proposes a novel approach based on deep learning for estimating ego motion with a downward looking camera. The network can be trained completely unsupervised and is not restricted to a specific motion model. We propose two neural network architectures based on the Early Fusion and Slow Fusion design principle: “EarlyBird” and “SlowBird”. Both networks share a Spatial Transformer layer for image warping and are trained with a modified structural similarity index (SSIM) loss function. Experiments carried out in simulation and for a real world differential drive robot show similar and partially better results of our proposed deep learning based approaches compared to a state-of-the-art method based on fast Fourier transformation.

show abstract

“…The advantages and challenges of event-based vision are well explained by the original work of Brandli et al [1] as well as the recent survey by Gallego et al [2]. The present work addresses intrinsic calibration of event cameras, a fundamental problem affecting potential future use in 3D vision applications such as [3], [4], [5], [6], [7], [8], [9], [10].…”

Section: Introductionmentioning

confidence: 99%

Dynamic camera calibration

Huang

Mitchell

Proceedings of International Symposium on Computer Vision - ISCV

View full text Add to dashboard Cite

Camera calibration is an important prerequisite towards the solution of 3D computer vision problems. Traditional methods rely on static images of a calibration pattern. This raises interesting challenges towards the practical usage of event cameras, which notably require image change to produce sufficient measurements. The current standard for event camera calibration therefore consists of using flashing patterns. They have the advantage of simultaneously triggering events in all reprojected pattern feature locations, but it is difficult to construct or use such patterns in the field. We present the first dynamic event camera calibration algorithm. It calibrates directly from events captured during relative motion between camera and calibration pattern. The method is propelled by a novel feature extraction mechanism for calibration patterns, and leverages existing calibration tools before optimizing all parameters through a multi-segment continuous-time formulation. As demonstrated through our results on real data, the obtained calibration method is highly convenient and reliably calibrates from data sequences spanning less than 10 seconds.

show abstract

Globally-Optimal Event Camera Motion Estimation

Cited by 22 publications

References 36 publications

Accuracy and Speed Improvement of Event Camera Motion Estimation Using a Bird’s-Eye View Transformation

Accuracy and Speed Improvement of Event Camera Motion Estimation Using a Bird’s-Eye View Transformation

Unsupervised deep learning based ego motion estimation with a downward facing camera

Dynamic camera calibration

Contact Info

Product

Resources

About