Multi-objective retrieval of object pose from video

Avanaki, Alireza Nasiri; Hamidzadeh, Babak; Kossentini, F.

doi:10.1109/tai.2000.889877

Cited by 4 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some pose estimation methods assume that depth map information is available for one frame [3] or all frames in the input sequence [13]. Although these can be used for the pose estimation of a general object, these methods fall into the model-based category since the depth map is actually information about the 3-D structure of the object.…”

Section: Related Researchmentioning

confidence: 99%

“…In this work, motivated by the fact that feature points are not always available and detectable, we put a 3-D reconstruction method (shape-from-silhouette for simplicity) and a model-based pose estimation method [3] in a volume feedback loop (Fig. 2) preceded by an initial pose estimation step (the technique specified in [3] fed by a flat depth map).…”

Section: Related Researchmentioning

confidence: 99%

“…2) preceded by an initial pose estimation step (the technique specified in [3] fed by a flat depth map). This system proves to be functional (Sect.…”

Section: Related Researchmentioning

confidence: 99%

“…The method we use for the RPE in this paper is very similar to the method introduced in [3] for the pose estimation of a rigid body in several frames. The method in [3] neither works with feature points, or regions, nor assumes a model for the object.…”

Section: Relative Pose Estimationmentioning

confidence: 99%

“…Initially, we assume flat depth maps for all frames. We then apply the method introduced in [3] to derive a set of estimated poses for all frames. Using this set of poses, we reconstruct the object volume, which in turn provides new depth information for all frames.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Multireference object pose retrieval with volume feedback

2004

View full text Add to dashboard Cite

Pose retrieval of a rigid object from monocular video sequences or images is addressed. Initially, the object pose is estimated in each image assuming flat depth maps. Shape-from-silhouette is then applied to make a 3-D model (volume), which is used for a new round of pose estimations, this time by a model-based method that gives better estimates. Before repeating this process by building a new volume, pose estimates are adjusted to reduce error by maximizing a novel quality factor for shape-from-silhouette volume reconstruction. The feedback loop is terminated when pose estimates do not change much, as compared with those produced by the previous iteration. Based on a theoretical study of the proposed system, a test of convergence to a given set of poses is devised. Reliable performance of the system is also proved by several experiments on both synthetic and real image sequences. No model is assumed for the object and no feature point is detected or tracked as there is no problematic feature matching or correspondence. Our method can be used for 3-D object tracking in video, 3-D modeling, and volume reconstruction from video.

show abstract

Section: Related Researchmentioning

confidence: 99%

Section: Related Researchmentioning

confidence: 99%

“…2) preceded by an initial pose estimation step (the technique specified in [3] fed by a flat depth map). This system proves to be functional (Sect.…”

Section: Related Researchmentioning

confidence: 99%

Section: Relative Pose Estimationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Multireference object pose retrieval with volume feedback

2004

View full text Add to dashboard Cite

show abstract

Object reconstruction and pose indexing by volume feedback

Avanaki

Hamidzadeh

Kossentini

Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429)

Self Cite

View full text Add to dashboard Cite

Three dimensional reconshuction of a rigid object from monocular video sequences is addressed. Initially object pose is estimated in each image by locating similar (unknown) textures assuming flat depth maps for all input images. Shape-from-silhouette [I] is then applied to make a 3-D model (volume), which is used for a new round of pose estimation, this time by a model-based method giving better estimates. Before repeating this process by building a new volume, pose estimates are adjusted to reduce enor by maximizing a quality measure fur shape-from-silhouette volume reconstruction. The volume feedback is terminated when pose estimates do not change much as compared to those produced by previous iteration. The final output is a pose index (the last set of pose estimates) and a volume. Good performance of the system is shown by several experiments. No model is assumed for the object. Feature points are neither detected nor tracked no problematic feahup matching or correspondence. The high-level pose index generated for input images can be used for content-based retrieval. Our method can be also applied to 3-D object tracking in video.

show abstract