Improving Human Computer Interaction 1461 Background

Extensive research has been invested in recent years into improving interactive segmentation algorithms. It is, however, striking that the human-computer interface, a substantial part of an interactive setup, is usually not investigated. Although the need for understanding the influence of human-computer interaction on interactive segmentation is recognized, only very little research has been done in this direction.

In order to improve information flow and to achieve optimal cooperation between interactive image analysis algorithms and human operators, we evaluated closed-loop systems utilizing new man-machine interfacing paradigms [62].

The mouse-based, manual initialization of deformable models in two dimensions represents a major bottleneck in interactive segmentation. In order to overcome the limitations of 2-D viewing and interaction the usage of direct 3-D interaction is inevitable. However, adding another dimension to user interaction causes several problems. Editing, controlling, and interacting in three dimensions often overwhelms the perceptual powers of a human operator. Furthermore, today's desktop metaphors are based on 2-D interaction and cannot easily be extended to the volumetric case. Finally, the visual channel of the human sensory system is not suitable for the perception of volumetric data.

However, these major drawbacks are valid only in terms of interactive systems that are based on 2-D Window-Icon-Mouse-Pointer (WIMP) interfaces that solely rely on the visual sense of the human operator. In order to alleviate the limitations of visual-only systems we may try to enhance the interaction process with additional sensory feedback. The fundamental challenge here is to find efficient ways for information flow between user and computer. Several sensory channels could be addressed, but due to the 3-D nature of the problem, the most obvious choice is the haptic channel. As an example, a multimodal system using visual and haptic volumetric rendering will be described, which was successfully applied to the segmentation of the intestinal system (Fig. 14.17).

Figure 14.17: Interactive, multimodal setup.
0 0

Post a comment