Although motorcycles are the main mode of transportation in many megacities in developing countries, very little research has been conducted on motorcycle operations and related issues. This paper addresses (1) the empirical analysis of the current status of motorcycle ownership and utilization in Ho Chi Minh City, (2) the estimation of the number of motorcycles used in this city, and (3) the analysis and modeling of different transportation policy scenarios. For (1) and (2), the survey results indicate that on average motorcycle ownership (2.33 motorcycles) is higher than that of working-age people in a family (2.29 people). The total number of motorcycles operating in Ho Chi Minh City is estimated to be almost 2 million fewer than that of registered motorcycles. For (3), the results show that each proposed scenario has its own advantage. The demand management policy, which involves either reducing demand via ride sharing or reducing the number of working days per fortnight to 9, is found to be the only single policy that results in little to no adverse impact on total vehicle trips, emissions, fuel consumption, or revenue.
Interactions between targets have been exploited to solve the occlusion problem in multitarget tracking but not to provide higher level scene understanding. As indoor environments are relatively unconstrained than urban areas, interactions in indoor environments are weaker and have more variants. Weak interactions make scene interaction modeling and neighboring object interaction modeling challenging. In this paper, a place-driven scene interaction model is proposed to represent long-term interactions in indoor environments. To deal with complicated short-term interactions, the neighboring object interaction model consists of three short-term interaction models, following, approaching and avoidance. The moving model, the stationary process model and these two interaction models are integrated to accomplish weakly interacting object tracking. In addition, higher level scene understanding such as unusual activity recognition and important place identification is accomplished straightforwardly. The experimental results using data from a laser scanner demonstrate the feasibility and robustness of the proposed approaches.