Showing papers by "Ishwar K. Sethi published in 2000"
04 Nov 2000
TL;DR: The system relies on an omni-face detection system capable of locating human faces over a broad range of views in color images or videos with complex scenes, which uses the presence of skin-tone pixels coupled with shape, edge pattern and face-specific features to locate faces.
Abstract: An omni-face detection scheme for image/video content description is proposed in this paper. It provides the ability to extract high-level features in terms of human activities rather than low-level features like color, texture and shape. The system relies on an omni-face detection system capable of locating human faces over a broad range of views in color images or videos with complex scenes. It uses the presence of skin-tone pixels coupled with shape, edge pattern and face-specific features to locate faces. The main distinguishing contribution of this work is being able to detect faces irrespective of their poses, including frontal-view and side-view, whereas contemporary systems deal with frontal-view faces only. The other novel aspects of the work lie in its iterative candidate filtering to segment objects from extraneous region, the use of Hausdorff distance-based normalized similarity measure to identify side-view facial profiles, and the exploration of hidden Markov model (HMM) to verify the presence of a side-view face. Image and video can be assigned with semantic descriptors based on human face information for later indexing and retrieval.
23 citations
TL;DR: From the experiments, the results of the neural network tested are similar to those given by the experienced doctors and better than those of previous research, indicating that this approach is very practical and beneficial to doctors comparing with some other methods currently existing.
Abstract: In this paper, we do some pre-processing on the input data to remove some noise before putting them into the network and some post-processing before outputting the results. Different neural networks such as back-propagation, radias basis network with different architecture are tested. We choose the one with the best performance among them. From the experiments we can see that the results of the neural network are similar to those given by the experienced doctors and better than those of previous research, indicating that this approach is very practical and beneficial to doctors comparing with some other methods currently existing.
19 Apr 2000
TL;DR: This paper considers a new method for object tracking which uses the information only supplied in compressed domain through the MPEG encoder, and results obtained indicate that tracking object under compressed domain is very promising.
Abstract: Currently, most approaches for object tracking are under spatial domain using optical flow and depth, or some model- based methods which need to uncompress those video sequences then do further disposal. The computation for uncompressing is expensive and not good for real time control. Although there are also some researchers doing object tracking under compressed domain, they only use part of DCT values of I frames in a video sequence, which doesn't take advantage of the information under compressed domain. In this paper we consider a new method for object tracking which uses the information only supplied in compressed domain through the MPEG encoder. The main scheme we have performed is to get the motion vectors of P and B frames directly from the MPEG video without decompressing it then cluster objects based on the motion vectors. In particular, camera motion is also taken into account since camera's motion can influence the objects' motion and segmentation results dramatically. Experiments based on the method mentioned above have been carried out in several videos. The results obtained indicate that tracking object in compressed domain is very promising.