A novel sequence representation for unsupervised analysis of human activities

Hamid, Raffay; Maddi, Siddhartha; Johnson, Amos; Bobick, Aaron F.; Essa, Irfan; Isbell, Charles L.

doi:10.1016/j.artint.2009.05.002

Cited by 98 publications

(47 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…There is yet no widely accepted methodology to recognize longer-term behaviours in complex scenes where obstructions and track breaks are common. The temporal structure can also be exploited by simpler approaches, where recent history is encoded in the representation itself [8,12]. We include this notion that recent history is an important queue as described in Section 4.4.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Instantaneous threat detection based on a semantic representation of activities, zones and trajectories

Burghouts¹,

Schutte²,

Hove³

et al. 2014

SIViP

View full text Add to dashboard Cite

Threat detection is a challenging problem, because threats appear in many variations and differences to normal behaviour can be very subtle. In this paper, we consider threats on a parking lot, where theft of a truck's cargo occurs. The theft takes place in very different forms, in the midst of many people who pose no threat. The threats range from explicit, e.g., a person attacking the truck driver, to implicit, e.g., somebody loitering and then fiddling with the exterior of the truck in order to open it. Our goal is a system that is able to recognize a threat instantaneously as they develop. Typical observables of the threats are a person's activity, presence in a particular zone, and the trajectory. The novelty of this paper is an encoding of these threat observables in a semantic, intermediate-level representation, based on low-level visual features that have no intrinsic semantic meaning themselves. The semantic representation encodes the notions of trajectories, zones and activities. The aim of this representation is to bridge the semantic gap between the low-level tracks and motion and the higher-level notion of threats. In our experiments, we demonstrate that our semantic representation is more descriptive for threat detection than directly using low-level features. We find that a person's activities are the most important elements of this semantic representation, followed by the person's trajectory. The proposed threat detection system is very accurate: 96.6% of the tracks are correctly interpreted, when considering the temporal context.

show abstract

Section: Related Workmentioning

confidence: 99%

“…An important design choice is whether the system is constructed by manual design [9,10,13] or trained [8,12,16]. The advantage of hand-crafted models is that expert knowledge can be included and that limited or no training data is required [9].…”

Section: Related Workmentioning

confidence: 99%

Instantaneous threat detection based on a semantic representation of activities, zones and trajectories

Burghouts¹,

Schutte²,

Hove³

et al. 2014

SIViP

View full text Add to dashboard Cite

show abstract

“…But the method requires activities to have time order constraints. Also [6] merges the scene POIs and sensorial information. But the method requires a manual specification of the scene.…”

Section: Related Workmentioning

confidence: 99%

Unsupervised Discovery, Modeling, and Analysis of Long Term Activities

Pusiol

Brémond

Thonnat

2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This work proposes a complete framework for human activity discovery, modeling, and recognition using videos. The framework uses trajectory information as input and goes up to video interpretation. The work reduces the gap between low-level vision information and semantic interpretation, by building an intermediate layer composed of Primitive Events. The proposed representation for primitive events aims at capturing meaningful motions (actions) over the scene with the advantage of being learned in an unsupervised manner. We propose the use of Primitive Events as descriptors to discover, model, and recognize activities automatically. The activity discovery is performed using only real tracking data. Semantics are added to the discovered activities (e.g., "Preparing Meal", "Eating") and the recognition of activities is performed with new datasets.

show abstract

“…to one of the 3 base events: Stationary (S), Walking (W) and Others (O), using variance as the main feature (see Figure 5b). We then form a sequence based representation (similar to that in [10]) of the user's activities while the user is in the coffee shop. Then, our classifier look for the leaving-from-the-line point which is usually the point where user performs at least 15 sec.…”

Section: Recognitionmentioning

confidence: 99%

LineKing: Coffee Shop Wait-Time Monitoring Using Smartphones

Bulut

Demirbaş

Ferhatosmanoğlu

2015

IEEE Trans. on Mobile Comput.

View full text Add to dashboard Cite

This article describes LineKing, a crowdsensing system for monitoring and forecasting coffee shop line wait times. LineKing consists of a smartphone component that provides automatic and accurate wait-time detection, and a cloud backend that uses the collected data to provide accurate wait-time estimation. LineKing is used on a daily basis by hundreds of users to monitor the wait-times of a coffee shop in the University at Buffalo, SUNY. The novel wait-time estimation algorithms of LineKing deployed at the cloud backend provide median absolute errors of less than 3 minutes. © 2002-2012 IEEE

show abstract

A novel sequence representation for unsupervised analysis of human activities

Cited by 98 publications

References 30 publications

Instantaneous threat detection based on a semantic representation of activities, zones and trajectories

Instantaneous threat detection based on a semantic representation of activities, zones and trajectories

Unsupervised Discovery, Modeling, and Analysis of Long Term Activities

LineKing: Coffee Shop Wait-Time Monitoring Using Smartphones

Contact Info

Product

Resources

About