Content deleted Content added
mNo edit summary |
mNo edit summary |
||
Line 16:
[[Image:Kitchen.png|right|thumb|
Figure 1. Layout for kitchen. (a) the “food preparation first” and (b) the “washing dishes first” meal preparation sequences. <br>
Image from <ref name="RecognitionOfHumanActivityThroughHierarchicalStochasticLearning" />]]
'''Hierarchical Hidden Markov Models (HHMMs)''' have been used in the context of eldercare aiming to identify behaviors such as eating dinner, watching television, and cooking <ref name="RecognitionOfHumanActivityThroughHierarchicalStochasticLearning" />. The goal is to learn the pattern of executing such activities normally, so that when an abnormal execution occurs (e.g., an elder person falls down while cooking), it can be automatically and instantaneously detected. Then, the appropriate action (e.g., provide audio/video cues to help the elderly or notify the appropriate caretaker that the elderly needs help) can then be taken before negative consequences for the elder person's health occur.
[[Supervised learning]] (a class of [[machine learning]] techniques) was used in <ref name="RecognitionOfHumanActivityThroughHierarchicalStochasticLearning" /> to learn an HHMM for four simplistic ways for dinner preparation ("spending some time preparing the food and rummaging through the fridge before the meal is cooked", "washing dishes prior to cooking on the stove", "going to the sink to wash the dishes then spending time at the food preparation area and the fridge before finally cooking the meal", "transition between each area of interest in a round robin fashion, starting and ending at the stove, before leaving the room") and five typical living room activities ("watch television", "read a book on the couch", "eat dinner", "eat dinner while watching television" and "there is nothing good on TV, read a book instead"). The person's position in the room was captured at short intervals of time and a sequence of observations was thus created. Observation sequences were labelled with the corresponding behavior and were used to learn HHMMs for the high-level dinner preparation and living room behaviors. Two example kitchen behaviors are shown in Figure 1. Figure 2 shows an HHMM learned for the "food preparation first" kitchen behavior.
[[Image:LearnedHHMM.png|
Figure 2. Learned HHMM for "food preparation first" kitchen behavior. <br>
Image from <ref name="RecognitionOfHumanActivityThroughHierarchicalStochasticLearning" />]]
Line 30:
Once the HHMMs for the kitchen and living room behaviors have been learned, they can be used to classify new sequences of observations. The experiments with recognizing simplistic kitchen and living room behaviors in <ref name="RecognitionOfHumanActivityThroughHierarchicalStochasticLearning" /> report high classification accuracy--83% for the cooking sequences and 80% for the living room sequences.
'''Abstract Hidden Markov Models (AHMM)''' have been used successfully to recognize human behavior based on the human's position in a building as recorded by video cameras<ref name="AHMM"/>. For instance, if the person is close to a computer, the behavior recognition method can infer that the person is using the computer, or if the person is moving in a hallway towards the north exit of the building, the method infers that the person intends to exit the building through the north exit.
== Advantages of Hierarchical Markov Models for Behavior Recognition ==
The ability of Hierarchical Markov Models to capture hierarchical decomposition makes them an appealing vehicle for capturing the hierarchical nature of human activities. It also allows for model re-usability <ref name="HierarchicalLearningAndPlanningInPOMDPs"> G. Theocharous [http://dl.acm.org/citation.cfm?id=936140 Hierarchical Learning and Planning in Partially Observable Markov Decision Processes]. PhD thesis, 2002. </ref> and certain conditional independence properties between different levels of abstraction in the model allow for faster learning and inference <ref name="AHMM"/> <ref name="RecognitionOfHumanActivityThroughHierarchicalStochasticLearning" />.
== References ==
| |||