Extracting Moving People from Internet Videos

Abstract. We propose a fully automatic framework to detect and extract arbitrary human motion volumes from real-world videos collected from YouTube. Our system is composed of two stages. A person detector is first applied to provide crude information about the possible locations of humans. Then a constrained clustering algorithm groups the detections and rejects false positives based on the appearance similarity and spatiotemporal coherence. In the second stage, we apply a top-down pictorial structure model to complete the extraction of the humans in arbitrary motion. During this procedure, a density propagation technique based on a mixture of Gaussians is employed to propagate temporal information in a principled way. This method reduces greatly the search space for the measurement in the inference stage. We demonstrate the initial success of this framework both quantitatively and qualitatively by using a number of YouTube videos.

• Type: Conference
• Year: 2008
• Where: ECCV
• Authors: Juan Carlos Niebles, Bohyung Han, Andras Ferencz, Fei-Fei Li 0002
• Keywords: Arbitrary Human Motion, Arbitrary Motion, Computer Vision, Constrained Clustering Algorithm, ECCV 2008, Rejects False Positives, Top-down Pictorial Structure