First Person Vision for Activity Prediction Using Probabilistic Modeling
Journal Title: Mehran University Research Journal of Engineering and Technology - Year 2018, Vol 37, Issue 4
Abstract
Identifying activities of daily living is an important area of research with applications in smart-homes and healthcare for elderly people. It is challenging due to reasons like human self-occlusion, complex natural environment and the human behavior when performing a complicated task. From psychological studies, we know that human gaze is closely linked with the thought process and we tend to “look” at the objects before acting on them. Hence, we have used the object information present in gaze images as the context and formed the basis for activity prediction. Our system is based on HMM (Hidden Markov Models) and trained using ANN (Artificial Neural Network). We begin with extracting motion information from TPV (Third Person Vision) streams and object information from FPV (First Person Vision) cameras. The advantage of having FPV is that the object information forms the context of the scene. When context is included as input to the HMM for activity recognition, the precision increases. For testing, we used two standard datasets from TUM (Technische Universitaet Muenchen) and GTEA Gaze+ (Georgia Tech Egocentric Activities). In the first round, we trained our ANNs only with activity information and in the second round added the object information as well. We saw a significant increase in the precision (and accuracy) of predicted activities from 55.21% (respectively 85.25%) to 77.61% (respectively 93.5%). This confirmed our initial hypothesis that including the focus of attention of the actor in the form of object seen in FPV can help in predicting activities better.
Authors and Affiliations
Shaheena Noor, Vali Uddin
Keywords
Related Articles
- EP ID EP394637
- DOI 10.22581/muet1982.1804.09
- Views 105
- Downloads 0
Urban Land Use Planning Trend and Sustainable Challenges in Socio-Economic Development
Land use planning is a technical approach for developing and managing the land into various public interests to endorse sustainable socio-economic development. This paper focuses on socio-economic problems by improper al...
LabVIEW Based Simulator for Solar Cell Characteristics and MPPT Under Varying Atmospheric Conditions
Though intermittent, solar energy is a clean and eternal source of energy. PV (Photovoltaic) cell is one of the technology to harness the solar energy and use it as electricity. In recent years rising cost of electricity...
Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands
A detailed laboratory study was carried out to investigate the controls and differences between Klinkenbergcorrected gas and liquid permeability. For this reason outcrop core samples were collected from Kirthar fold belt...
M-Health Service for Train Passengers Using Mobile GPS System: An ArchiMate Service Layer Model
EA (Enterprise Architecture) is an instrument that is employed to describe the organization?s structure, business layout and operations within the IT (Information Technology) environment. Different types of organizations...
Statistical Modeling for Stability of Emulsion Liquid Membrane for the Removal of Anionic Dyes from Textile Wastewater
The importance of statistical modeling is obvious because it imparts an imperative role in predicting the effects of significant factors on any experimental process. In recent days, ELMs (Emulsion Liquid Membranes) are c...