Probabilistic 3D Tracking: Rollator Users' Leg Pose from Coronal Images

Ng, Samantha; Fakih, Adel; Fourney, Adam; Poupart, Pascal; Zelek, John

doi:10.1109/crv.2009.45

Cited by 1 publication

(8 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As mentioned in [13], the cues are not actully independent of one another given a hypothesis. We are exploring statistical dependencies between texture and colour using methods described in [18].…”

Section: B Combining Weightsmentioning

confidence: 96%

“…We compare the performance of specific colour and texture models to the four general cues in [13]. The general cues are summarized as follows (see [13] for implementation details).…”

Section: A Appearance Model Evaluationmentioning

confidence: 99%

“…The general cues are summarized as follows (see [13] for implementation details). First and second, assuming homogeneity within these regions, then leg regions are characterized by a very low average gradient magnitude and uniformity in colour.…”

Section: A Appearance Model Evaluationmentioning

confidence: 99%

“…Third is the expected symmetry that people tend to exhibit in their left and right body segments. While symmetry is assumed in our initialization of the appearance, it is explicitly evaluated in [13]. Fourth is the expected contrast between the foot and an uncluttered floor.…”

Section: A Appearance Model Evaluationmentioning

confidence: 99%

“…To our knowledge, there is no system that adequately addresses all of the constraints of our rollator application. [13] explored a general appearance model, based on multiple cues, that required no detection in an initialization phase. Here we explore the advantages of using a more specific appearance model based on texture and colour, which minimizes our reliance on heuristics.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Towards a mobility diagnostic tool: Tracking rollator users' leg pose with a monocular vision system

Fakih

Fourney

et al. 2009

2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society

View full text Add to dashboard Cite

Abstract-Cognitive assistance of a rollator (wheeled walker) user tends to reduce the attentional capacity of the user and may impact her stability. Hence, it is important to understand and track the pose of rollator users before augmenting a rollator with some form of cognitive assistance. While the majority of current markerless vision systems focus on estimating 2D and 3D walking motion in the sagittal plane, we wish to estimate the 3D pose of rollator users' lower limbs from observing image sequences in the coronal (frontal) plane. Our apparatus poses a unique set of challenges: a single monocular view of only the lower limbs and a frontal perspective of the rollator user. Since motion in the coronal plane is relatively subtle, we explore multiple cues within a Bayesian probabilistic framework to formulate a posterior estimate for a given subject's leg limbs. In this work, our focus is on evaluating the appearance model (the cues). Preliminary experiments indicate that texture and colour cues conditioned on the appearance of a rollator user outperform more general cues, at the cost of manually initializing the appearance offline.

show abstract

Section: B Combining Weightsmentioning

confidence: 96%

“…We compare the performance of specific colour and texture models to the four general cues in [13]. The general cues are summarized as follows (see [13] for implementation details).…”

Section: A Appearance Model Evaluationmentioning

confidence: 99%

Section: A Appearance Model Evaluationmentioning

confidence: 99%

Section: A Appearance Model Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations