Recognition of repetitive sequential human activity

Fan, Quanfu; Bobbitt, Russell; Zhai, Yun; Yanagawa, Akira; Pankanti, Sharath; Hampapur, Arun

doi:10.1109/cvpr.2009.5206644

Cited by 25 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As expected, PHF has a lower precision than STIP , which is roughly 4% less on average on ST-1 and 3% less on ST-2. However, when combining the primitive detection results into visual scans using the approach in [4], it leads to an even smaller performance difference less than 2%, clearly showing that PHF is a qualified substitute for STIPs in the retail context.…”

Section: Event Detection Resultsmentioning

confidence: 85%

“…Evaluation Measure We evaluated our algorithms in the same way as in [4]. We define the overlap percentage of a primitive e 1 and a prediction e 2 as their intersection divided by the union of the two events, i.e, τ = e 1 take the one with the maximum overlap percentage as the correct match if the percentage exceeds some threshold τ .…”

Section: Resultsmentioning

confidence: 99%

“…Our Approach Our main focus of this paper is to develop efficient primitive detectors to make practical use of the Constrained Viterbi algorithm proposed in [4] for fraud detection. Motivated by a vision-based touch detection algorithm [5] designed for user interfaces, we simulate the cashier's pickup and drop actions as a button-touch-like interaction with the surface of the belt.…”

Section: Introductionmentioning

confidence: 99%

“…To help retailers reduce retail loss, we proposed a novel video analytic approach in [4] to detect sweethearting at the POS based on a Constrained Viterbi algorithm. The approach is a general probabilistic framework for grouping primitives with strong temporal dependencies into repetitive sequential events such as visual scans discussed here.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Fast detection of retail fraud using polar touch buttons

Fan

Yanagawa

Bobbitt

et al. 2009

2009 IEEE International Conference on Multimedia and Expo

Self Cite

View full text Add to dashboard Cite

Video analytics have recently emerged as a promising technique of retail fraud detection for loss prevention. Efficient video analytic algorithms are highly desired for a practical fraud detection system. In this paper, we present a real-time algorithm for recognizing a cashier's actions at the Point of Sale (POS), which can be further used to analyze cashier behaviors for identifying fraudulent incidents. The algorithm uses a set of simple but effective features derived from a global representation of motion energy called Polar Motion Map (PMM). These features capture the motion patterns exhibited in a cashier's actions as a focused beam of motion energy, characterizing the actions as the extension and retraction movement of the cashier's arm with respect to a prespecified region. Our algorithm demonstrates comparable accuracy against one of the state-of-the-art event recognition techniques [1] while running significantly faster.

show abstract

Section: Event Detection Resultsmentioning

confidence: 85%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fast detection of retail fraud using polar touch buttons

Fan

Yanagawa

Bobbitt

et al. 2009

2009 IEEE International Conference on Multimedia and Expo

Self Cite

View full text Add to dashboard Cite

show abstract

“…Messing et al [55] used the velocity history of tracked key-points to represent actions and then combined other features, such as appearance, position, and high level sematic features together to recognize activities of daily living. In [56], Fan et al modified the Viterbi algorithm to detect repetitive sequential event units and then recognized the predominant retail cashier activities at the checkout station.…”

Section: Related Workmentioning

confidence: 99%