Comparison of Point and Line Features and Their Combination for Rigid Body Motion Estimation

Pilz, Florian; Pugeault, Nicolas; Krüger, Norbert

doi:10.1007/978-3-642-03061-1_14

Cited by 8 publications

(11 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The mathematical formulation of the RBM that we use is from (Rosenhahn et al, 2001), and has three advantages: First, the motion is optimised in 3D space; second, it allows for solving the motion jointly for different kind of constraint equations that stem from different type of image features (in this case, local edge descriptors and SIFT); third, it minimises the error directly in SE(3), and therefore does not require additional measures to handle degenerate cases. As been shown in (Pilz et al, 2009), a combination of heterogeneous features (edges and SIFT features) leads to an improved robustness and accuracy of the RBM estimate. Outliers are discarded using RANSAC (Fischler and Bolles, 1981).…”

Section: Rigid Body Motion (Rbm) Estimationmentioning

confidence: 73%

A Three-Level Architecture for Model–free Detection and Tracking of Independently Moving Objects

Pugeault

Pauwels

Hulle

et al. 2010

Proceedings of the International Conference on Computer Vision Theory and Applications

View full text Add to dashboard Cite

Abstract:We present a three-level architecture for detection and tracking of independently moving objects (IMOs) in sequences recorded from a moving vehicle. At the first stage, image pixels with an optical flow that is not entirely induced by the car's motion are detected by combining dense optical flow, egomotion extracted from this optical flow, and dense stereo. These pixels are segmented and an attention mechanism is used to process them at finer resolution at the second level making use of sparse 2D and 3D edge descriptors. Based on the rich and precise information on the second level, the full rigid motion for the environment and for each IMO is computed. This motion information is then used for tracking, filtering and the building of a 3D model of the street structure as well as the IMO. This multi-level architecture allows us to combine the strength of both dense and sparse processing methods in terms of precision and computational complexity, and to dedicate more processing capacity to the important parts of the scene (the IMOs).

show abstract

Section: Rigid Body Motion (Rbm) Estimationmentioning

confidence: 73%

A Three-Level Architecture for Model–free Detection and Tracking of Independently Moving Objects

Pugeault

Pauwels

Hulle

et al. 2010

Proceedings of the International Conference on Computer Vision Theory and Applications

View full text Add to dashboard Cite

show abstract

“…First, all primitives are tracked over time and filtered using an Unscented Kalman Filter based on the combination of prediction, observation and update stages. In the prediction stage, the system's knowledge of the motion (e.g., the motion of the robot arm) or an estimated motion (see e.g., [61]) is used to calculate the poses of all accumulated primitives at the next time step. The observation stage matches the predicted primitives with their newly observed counterparts.…”

Section: Task 4: Disambiguation Via Accumulationmentioning

confidence: 99%

Using multi-modal 3D contours and their relations for vision and robotics

Baseski

Pugeault

Kalkan

et al. 2010

Journal of Visual Communication and Image Representation

Self Cite

View full text Add to dashboard Cite

“…4(c-ii), the putative positions x (2,1) and x (3,1) , elicited by two distinct hexagonal cells, represent the same image location. Moreover, the filter's spatial extension can lead to proximate positions describing essentially the same image structure: in Fig.…”

Section: Elimination Of Redundant Descriptorsmentioning

confidence: 99%

“…6. Meanwhile, this descriptor has been used in a number of applications such as the learning of object representations, 1 pose estimation, 2 motion estimation, 3 and vision-based grasping. 4 In these applications, we observed the importance of three properties the edge descriptor fulfills: explicitness, orthogonality, and condensation.…”

Section: Introductionmentioning

confidence: 99%

Visual Primitives: Local, Condensed, Semantically Rich Visual Descriptors and Their Applications in Robotics

Pugeault

Wörgötter

Krüger

2010

Int. J. Human. Robot.

Self Cite

View full text Add to dashboard Cite

We present a novel representation of visual information, based on local symbolic descriptors, that we call visual primitives. These primitives: (1) combine different visual modalities, (2) associate semantic to local scene information, and (3) reduce the bandwidth while increasing the predictability of the information exchanged across the system. This representation leads to the concept of early cognitive vision that we define as an intermediate level between dense, signal-based early vision and high-level cognitive vision. The framework's potential is demonstrated in several applications, in particular in the area of robotics and humanoid robotics, which are briefly outlined.

show abstract

Comparison of Point and Line Features and Their Combination for Rigid Body Motion Estimation

Cited by 8 publications

References 37 publications

A Three-Level Architecture for Model–free Detection and Tracking of Independently Moving Objects

A Three-Level Architecture for Model–free Detection and Tracking of Independently Moving Objects

Using multi-modal 3D contours and their relations for vision and robotics

Visual Primitives: Local, Condensed, Semantically Rich Visual Descriptors and Their Applications in Robotics

Contact Info

Product

Resources

About