100 likes | 226 Views
Scene Understanding and Assisted Living Francois Bremond PULSAR, INRIA. 29/11/2010. Date. Monitoring of Activities of Daily Living for Elderly. Motivation : Increase independence and quality of life: Enable elderly to live longer in their preferred environment.
E N D
Scene Understanding and Assisted Living Francois BremondPULSAR, INRIA 29/11/2010 Date
Monitoring of Activities of Daily Living for Elderly • Motivation : Increase independence and quality of life: • Enable elderly to live longer in their preferred environment. • Reduce costs for public health systems. • Relieve family members and caregivers. • Technical Goals : • Detecting alarming situations (eg. Falls) • Assess the degree of frailty of elderly people. • Detecting changes in behavior (missing activities, disorder, interruptions, repetitions, inactivity). • Building a video library of reference behaviors characterizing people frailty. Example of a regular activity: Meal preparation (in kitchen) (11h– 12h) Eating (in dinning room) (12h -12h30) Resting, TV watching, (in living room) (13h– 16h) …
1st experiment : Gerhome laboratory • Partners:INRIA, CSTB, Nice Hospital, CG06…
Approach: Real time semantic activity understanding Posture Recognition Activity Recognition Detection Tracking Classification Actions Person inside Kitchen 4
Language combining multi-sensor information A language to model complex activities Activity (Use Fridge, Physical Objects ( (p: Person), (Fridge: Equipment), (Kitchen: Zone)) Components ((c1: Inside zone (p, Kitchen)) (c2: Close_to (p, Fridge)) (c3: Bending (p) (c4: Opening (Fridge)) (c5: Closing (Fridge)) ) Constraints ((c1 beforec2 ) (c3 duringc2 ) (c4:time + 10s < c5:time) )) Detected by video camera Detected by contact sensor 5
Event recognition results • Recognition of the “Having meal” event for a 84 old woman
Recognition of a set of activities comparing two elderly people Table 2: Monitored activities, their frequency (n1 & n2), mean duration (m1 & m2) and total duration for 2 volunteers staying in the GERHOME laboratory for 4 hours; NDA=Normalized Difference of mean durations of Activities=|m1-m2|/ (m1+m2); NDI=Normalized Difference of Instances number=|n1-n2|/(n1+n2); possible differences in behavior of the 2 volunteers are signified in bold
2nd experiment : CMRR in Nice Hospital We propose a new protocol to study Alzheimer disease. Recognition of the “stop and go” activity.
3rd experiment : Learning Scenario Models Localization of the person during 4 observation hours Stationerypositions of the person Walked distance = 3.71 km
Conclusion Contributions: operational monitoring system • Cognitive vision approach to recognize activities of elderly using video cameras and other sensors. • Model 24 activities related to ADLs of elderly (normal activities and abnormal situations) and validation on 9 elderly people • Future Work: improving robustness • Validate the performance • on more elderly people • with Alzheimer patients at Nice hospital. • Learning scenario models • -> Building a library of reference behaviors based on video