سوال ۱۴ کنکور ارشد علوم کامپیوتر ۱۴۰۴

PASSAGE 1 (excerpt):

Paragraph 1: Human-computer interaction (HCI) has been the focus of attention for researchers in the past decade, with considerable work being done in the various modalities for communicating with a computer. Many interfaces, relying on body gestures, speech and brain computing, have been proposed or developed which assist humans to interact with robots in a more intelligent and natural way. Abstract thoughts and commands can be communicated to such interfaces more easily, which thereafter processes them intelligently, using past heuristics and concepts.

Paragraph 2: It is increasingly apparent that no single input modality can completely represent human thought and commands accurately, because an individual gesture can represent different concepts depending upon the context in which it is used. Thus, recognition of human gestures is mostly an ill-posed problem, since there might not be a unique solution (concept) for a detected gesture. To combine or improve the recognition of the gesture during the understanding stage, multimodal interaction interfaces, which combine cues from different input modalities, allow for a better representation of human intent and facilitate interaction. In this respect, multimodal interaction interfaces, which combine cues from different input modalities, allow for a better representation of human intent and facilitate interaction.

Question: According to paragraph 2, the problem of recognition of human gestures can be improved by .................. .