Multidimensional adaptive sampling and reconstruction for ray tracing
… reconstruction by determining the extent of each sample in … only in the image plane poses
problems in the context of … Our approach performs adaptive sampling by placing samples near …
problems in the context of … Our approach performs adaptive sampling by placing samples near …
Adapting the sample size in particle filters through KLD-sampling
D Fox - The international Journal of robotics research, 2003 - journals.sagepub.com
… using the Kullback-Leibler distance. Our adaptation approach chooses a small number of
samples … space, and it chooses a large number of samples if the state uncertainty is high. Both …
samples … space, and it chooses a large number of samples if the state uncertainty is high. Both …
Visual tracking and recognition using appearance-adaptive models in particle filters
SK Zhou, R Chellappa… - … Transactions on Image …, 2004 - ieeexplore.ieee.org
… confronted by severe pose and illumination variations. We improve our approach in the
following … To downweigh the influence of the outlier pixels in estimating the adaptive velocity, we …
following … To downweigh the influence of the outlier pixels in estimating the adaptive velocity, we …
A method for the registration of attributed range images
G Godin, D Laurendeau… - … on 3-D Digital Imaging …, 2001 - ieeexplore.ieee.org
… the problem of registration from arbitrary initial relative poses, not just of refinement. It is …
: first their distribution orients the selection of random samples; secondly the determination …
: first their distribution orients the selection of random samples; secondly the determination …
A real time adaptive visual surveillance system for tracking low-resolution colour targets in dynamically changing scenes
P KaewTrakulPong, R Bowden - Image and Vision Computing, 2003 - Elsevier
… , our method is designed to track small colour targets commonly found in outdoor visual
surveillance. Our approach … to track people, identifying poses and separating individuals during …
surveillance. Our approach … to track people, identifying poses and separating individuals during …
Zippered polygon meshes from range images
… to the surface detail by finding the consensus geometry. The … of nearby positions from each
of the original range images. The … other integration methods that will correctly determine the …
of the original range images. The … other integration methods that will correctly determine the …
Real-time object recognition in sparse range images using error surface embedding
L Shang, M Greenspan - International Journal of Computer Vision, 2010 - Springer
… recognize and determine the pose of threedimensional (3-D) objects in range image data has
… We also propose a practical method to construct an effective generic model, and examine …
… We also propose a practical method to construct an effective generic model, and examine …
Sampling strategies for bag-of-features image classification
… owing to changes in pose, imaging and lighting conditions, … random sample of recent work
on local feature based methods… that are biased towards finding more discriminant patches. …
on local feature based methods… that are biased towards finding more discriminant patches. …
Adaptively sampled distance fields: A general representation of shape for computer graphics
SF Frisken, RN Perry, AP Rockwood… - … and interactive techniques, 2000 - dl.acm.org
… method for estimating surface normals from the sampled data. For … use adaptive sampling,
they should provide significant compression over LUT representations. Finally, since distance …
they should provide significant compression over LUT representations. Finally, since distance …
Parametrized structure from motion for 3d adaptive feedback tracking of faces
TS Jebara, A Pentland - Proceedings of IEEE Computer …, 1997 - ieeexplore.ieee.org
… Motion provides a strong cue for estimating 3D structure, pose and … , pose, illumination
and background clutter. We propose a system that uses simple image processing techniques to …
and background clutter. We propose a system that uses simple image processing techniques to …