Long-Range Trajectories from Global and Local Motion Representations

TL;DR

This paper introduces a novel motion-based system that encodes long-range trajectories by integrating local and global motion representations, improving scene analysis and human activity understanding across diverse scenarios.

Contribution

The paper presents a generic system formulation for long-range trajectories that handles scene variability and motion changes, advancing trajectory-based analysis methods.

Findings

Effective in various scenarios for human activity tasks

Improves trajectory robustness over traditional methods

Provides a new evaluation framework for motion analysis

Abstract

Motion is a fundamental cue for scene analysis and human activity understan- ding in videos. It can be encoded in trajectories for tracking objects and for action recognition, or in form of flow to address behaviour analysis in crowded scenes. Each approach can only be applied on limited scenarios. We propose a motion-based system that represents the spatial and temporal features of the flow in terms of long-range trajectories. The novelty resides on the system formulation, its generic approach to handle scene variability and motion variations, motion integration from local and global representations, and the resulting long-range trajectories that overcome trajectory-based approach problems. We report the results and conclusions that state its pertinence on different scenarios, comparing and correlating the extracted trajectories of individual pedestrians, manually annotated. We also propose an evaluation framework and stress the diverse system characteristics that can be used for human activity tasks, namely on motion segmentation.