Sorry, you need to enable JavaScript to visit this website.

In this paper, a multiscale image decomposition method based on domain transform is proposed. The domain transform is a high speed edge preserving smoothing method and can be used to many image processing applications. However, it is highly sensitive to noise. The proposed method is based on filters used in the domain transform but is designed to be robust to noise by employing a multiscale method. An optimization problem is formulated to obtain desired domain- transformed output. As expected, the method can be used to many applications as the domain transform.

In this work, we propose an improved fast multiple-view image denoising algorithm using 3D focus image stacks. It showed improved computational efficiency and comparable denoising quality compared to conventional methods.

Presentation slides covering:

- robust foreground detection / background subtraction via patch-based analysis
- person re-identification based on representations on Riemannian manifolds
- robust object tracking via Grassmann manifolds
- adapting the lessons from big data to computer vision
- future paradigm shifts: computer vision based on networks of neurosynaptic cores

Slides from the Tutorial on Riemannian Geometry in Computer Vision, presented at the Asian Conference on Computer Vision (ACCV), Singapore, 2014.

The slides show (1) how objects can be interpreted as points on Riemannian and Grassmann manifolds, and (2) various distance measures on manifolds. Demonstrates usefulness of manifold techniques in applications such as object tracking and person re-identification.

For the purposes of foreground estimation, the true background model is unavailable in many practical circumstances and needs to be estimated from cluttered image sequences. We propose a sequential technique for static background estimation in such conditions, with low computational and memory requirements. Image sequences are analysed on a block-by-block basis. For each block location a representative set is maintained which contains distinct blocks obtained along its temporal line.

Expanded version of the Guest Editorial
for Special Issue on Signal Processing for Art Investigation
(IEEE Signal Processing Magazine, July 2015)

Include short summaries for each of the 11 articles in the special issue.

We propose an adaptive tracking algorithm where the object is modelled as a continuously updated bag of affine subspaces, with each subspace constructed from the object's appearance over several consecutive frames. In contrast to linear subspaces, affine subspaces explicitly model the origin of subspaces. Furthermore, instead of using a brittle point-to-subspace distance during the search for the object in a new frame, we propose to use a subspace-to-subspace distance by representing candidate image areas also as affine subspaces.