Geometric particle swarm optimization for robust visual ego-motion estimation via particle filtering

Baik, Young Ki; Junghyun, Kwon; Lee, Hee-Seok; Lee, Kyoung Mu

doi:10.1016/j.imavis.2013.04.004

Cited by 7 publications

(5 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We validate the proposed method by simulations 1 and compare its performance to that of a state-of-the-art tracking algorithms presented in [9], [16] and [17]. These works use a standard with a constant velocity motion model, and therefore differ from our proposed method both in filtering technique and motion model.…”

Section: Numerical Validationmentioning

confidence: 96%

“…The velocity and pose at the end of the time-step are computed using ( 13) and ( 14), respectively. The values for dP N and dP T are found by solving the so-called contact problem, defined by the setvalued force laws of ( 16), (17), (18), and (19). We use an augmented Lagrangian approach as discussed in [29] to solve this contact problem, due to its simplicity and effectiveness in our specific case of a single rigid body.…”

Section: Numerical Integration Using Time-steppingmentioning

confidence: 99%

“…In [9], [16], and [17], a constant velocity model is used to describe the state transition on SE(3). In this constant velocity model, no preference is given to any direction of motion such that the equations of motion can be written as…”

Section: Numerical Validationmentioning

confidence: 99%

“…Consequently, it may happen that the complete set of particles moves away from the area with high likelihood, resulting in degeneracy of the particle set [15]. Second, in literature of visual object tracking, the adopted motion model is typically a constant velocity model [9], [16], [17]. To the best of the authors' knowledge, there does not exist literature on object tracking algorithms exploring the possibility of incorporating prior information regarding the object and surface that are expected to get in contact, potentially undergoing impacts (nonsmooth behavior).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Model-Based 6D Visual Object Tracking with Impact Collision Models

Jongeneel

Bernardino

Wouw

et al. 2022

2022 American Control Conference (ACC)

View full text Add to dashboard Cite

3D Object tracking is an essential technique in computer vision and has many fields of application. In this study, the focus lies on tracking objects impacting the environment. We show that state-of-the-art methods lose track of objects in this context and we investigate how to overcome this problem by adding prior information regarding the object and surface where collision is expected to occur. For illustration purposes and application relevance, we focus on the case of a box impacting a surface, which is, e.g., encountered in robot tossing in logistics applications. We model the effects of impacts and friction in a motion model, and consider the state of the box to evolve in a Lie group. We present an object tracking algorithm, based on an Unscented Particle Filter, for systems whose state lives in a Lie group and incorporate this motion model. The observations are taken from a single RGB camera and make use of the known 3D model of the object and color characteristics to predict its appearance in the 2D image. We quantitatively evaluate the effectiveness of our proposed methods by means of simulations on synthetic images.

show abstract

Section: Numerical Validationmentioning

confidence: 96%

Section: Numerical Integration Using Time-steppingmentioning

confidence: 99%

Section: Numerical Validationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Model-Based 6D Visual Object Tracking with Impact Collision Models

Jongeneel

Bernardino

Wouw

et al. 2022

2022 American Control Conference (ACC)

View full text Add to dashboard Cite

show abstract

“…There are, however, some bio-inspired metaheuristics that mimic the behavior of natural organisms (e.g. Genetic Algorithms (GAs) and Particle Swarm Optimization (PSO) [16]- [18]) or the physical laws of nature to cater this problem [19]. These methods have two common functionalities: exploration and exploitation.…”

Section: Introductionmentioning

confidence: 99%