Min Gyo Chung

Relevance feedback reinforced with semantics accumulation

Relevance feedback (RF) is a mechanism introduced earlier to exploit a user’s perceptual feedback in image retrieval. It refines a query by using the relevance information from the user to improve subsequent retrieval. However, the user’s feedback information is generally lost after a search session terminates. In this paper, we propose an enhanced version of RF, which is designed to accumulate human perceptual responses over time through relevance feedback and to dynamically combine the accumulated high-level relevance information with low-level features to further improve the retrieval effectiveness. Experimental results are presented to demonstrate the potential of the proposed method.

Estimating Face Direction via Facial Triangle

In this paper, we propose a vision-based approach to detect a face direction from a single monocular view of a face by using a facial feature called facial triangle. Specifically, the proposed method introduces formulas to detect face rotation, horizontally and vertically, using the facial triangle. Our method makes no assumption about the structure of the face and produces an accurate estimate of face direction.

Efficient Management of Domain Foreign Agents in Mobile Computing Environment Using Load Balance

Mobile IP is a protocol standard designed to be used in a mobile computing environment. However, Mobile IP has a drawback to incur a lot of handoff delays and waste network resources, since CoA registration packets need to go through a HA first whenever a mobile node moves. To solve such problem, this paper proposes a new scheme that, for intra-domain movement, efficiently performs local handoff without notifying the HA. Specifically, based on the notion of load balance, the proposed scheme allows every FA in a domain to become the root FA (a.k.a. Domain FA) dynamically, thus distributing the registration task into many other foreign agents. Our simulation results show that the proposed method proves to reduce registration packets by approximately 7–15% more than existing methods.