Abstract
We present a method for characterizing human activities using 3D deformable shape models. The motion trajectories of points extracted from objects involved in the activity are used to build models for each activity, and these models are used for classification and detection of unusual activities. The deformable models are learnt using the factorization theorem for nonrigid 3D models. We present a theory for characterizing the degree of deformation in the 3D models from a sequence of tracked observations. This degree, termed as deformation index (DI), is used as an input to the 3D model estimation process. We study the special case of ground plane activities in detail because of its importance in video surveillance applications. We present results of our activity modeling approach using videos of both high-resolution single individual activities and ground plane surveillance activities.
Highlights
Activity modeling and recognition from video is an important problem, with many applications in video surveillance and monitoring, human-computer interaction, computer graphics, and virtual reality
We show the effect of using a 3D model in recognizing these activities from different viewing angles
We have presented a framework for using 3D deformable shape models for activity modeling and representation
Summary
Activity modeling and recognition from video is an important problem, with many applications in video surveillance and monitoring, human-computer interaction, computer graphics, and virtual reality. The problem of activity modeling is associated with modeling a representative shape which contains significant information about the underlying activity This can range from the shape of the silhouette of a person performing an action to the trajectory of the person or a part of his body. In many of the 3D approaches, a 2D shape is represented by a finite-dimensional linear combination of 3D basis shapes and a camera projection model relating the 3D and 2D representations [7,8,9,10] This method has been applied primarily to deformable object modeling and tracking. In [11], actions under different variability factors were modeled as a linear combination of spatiotemporal basis actions The recognition in this case was performed using the angles between the action subspaces without explicitly recovering the 3D shape. This approach needs sufficient video sequences of the actions under different viewing directions and other forms of variability to learn the space of each action
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
