A real-time tracker for markerless augmented reality

Comport, Andrew I.; Marchand, Éric; Chaumette, F.

doi:10.1109/ismar.2003.1240686

Cited by 110 publications

(93 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In computer vision, the geometric primitives considered for tracking have been numerous, however, amongst them distance based features have shown to be efficient and robust [17,11,4]. Another important issue is the 2D-3D registration problem.…”

Section: State Of the Artmentioning

confidence: 99%

“…Linear approaches use a least-squares method to estimate the pose and are considered to be more suitable for initialization procedures. Full-scale non-linear optimization techniques (e.g., [17,19,11,4]) consist of minimizing the error between the observation and the forward-projection of the model. In this case, minimization is handled using numerical iterative algorithms such as Newton-Raphson or Levenberg-Marquardt.…”

Section: State Of the Artmentioning

confidence: 99%

“…This minimization can be seen as the dual problem of visual servoing whereby minimizing the parameters corresponds to controlling an arm-to-eye system so as to observe the arm at a given position in the image (note that an object is not necessarily fixed to the ground). This duality is known as Virtual Visual Servoing [4,20]. The advantages of the rigid body tracking method, presented in this paper, include:…”

Section: Contributionmentioning

confidence: 99%

“…These include manually making the correspondence of four point features with the model [7], a manual manipulation of the object projected on the image (using the mouse) or an automatic procedure in the case of textured objects [16]. The initial estimate is then refined using the Virtual Visual Servoing [4,20] non-linear iterative minimization approach. Following an approximate initialization procedure, the closed-loop tracking procedure proposed in this paper takes over.…”

Section: Overview and Motivationsmentioning

confidence: 99%

See 3 more Smart Citations

Kinematic sets for real-time robust articulated object tracking

Comport

Marchand

Chaumette

2007

Image and Vision Computing

View full text Add to dashboard Cite

In this article a new approach is given for real-time visual tracking of a class of articulated non-rigid objects in 3D. The main contribution of this paper consists in symmetrically modeling the motion and velocity of an articulated object via a novel kinematic set approach. This is likened to a Lagrange-d'Alembert formulation in classical physics. The advantages of this new model over pre-existing methods include improved precision, robustness and efficiency, leading to real-time performance. Furthermore, a general class of mechanical joints can be considered and the method can track objects where previous approaches have failed due to a lack of visual information. In summary, a joint configuration is modeled by using Pfaffian velocity constraints. The configuration and location of a joint is then used to build a general Jacobian Matrix which relates individual rigid body velocities(twists) to an underlying minimal subspace. A closed loop control law is then derived in order to minimize a set of distance errors in the image and estimate the system parameters. The tracking is locally based upon efficient distance criterion. Experimental results show prismatic, rotational and helical type links and up to eight general parameters. A statistical M-estimation technique is applied to improve robustness. A monocular camera system was used as a real-time sensor to verify the theory.

show abstract

Section: State Of the Artmentioning

confidence: 99%

Section: State Of the Artmentioning

confidence: 99%

Section: Contributionmentioning

confidence: 99%

Section: Overview and Motivationsmentioning

confidence: 99%

See 2 more Smart Citations

Kinematic sets for real-time robust articulated object tracking

Comport

Marchand

Chaumette

2007

Image and Vision Computing

View full text Add to dashboard Cite

show abstract

“…In the bagof-features approach, local coordinate frames are constructed around points of interest [5], [9], and features from each local frame vote for a certain object and pose hypothesis. In the model-based approach [2], [11], a geometrical model is fitted to the observed image. This approach is often very accurate, but requires a good initial guess and a manually constructed 3D model.…”

Section: Introductionmentioning

confidence: 99%

Accurate Interpolation in Appearance-Based Pose Estimation

Felsberg

Image Analysis

View full text Add to dashboard Cite

Abstract. One problem in appearance-based pose estimation is the need for many training examples, i.e. images of the object in a large number of known poses. Some invariance can be obtained by considering translations, rotations and scale changes in the image plane, but the remaining degrees of freedom are often handled simply by sampling the pose space densely enough. This work presents a method for accurate interpolation between training views using local linear models. As a view representation local soft orientation histograms are used. The derivative of this representation with respect to the image plane transformations is computed, and a Gauss-Newton optimization is used to optimize all pose parameters simultaneously, resulting in an accurate estimate.

show abstract