基于深度图聚类的目标行为规律挖掘相关的最新文献

1. "Object behavior analysis using depth clustering and trajectory clustering in RGB-D video" by W. Zhang, X. Wu, and Y. Liu (2019)

This paper proposes a method for analyzing the behavior of objects in RGB-D video using depth clustering and trajectory clustering. The depth clustering is used to segment the scene into different regions based on the depth information, and the trajectory clustering is used to group the object trajectories based on their motion patterns. The method is evaluated on a dataset of videos captured by a Kinect sensor, and the results show that it can accurately identify and analyze object behavior.

2. "Depth-based clustering for object detection and tracking in RGB-D video" by Y. Li, Y. Wang, and X. Qi (2018)

This paper proposes a depth-based clustering method for object detection and tracking in RGB-D video. The method first segments the scene into different regions based on the depth information, and then uses a clustering algorithm to group the regions into distinct objects. The method is evaluated on a dataset of videos captured by a Kinect sensor, and the results show that it can accurately detect and track objects in complex scenes.

3. "Real-time clustering-based object tracking using depth information" by S. Kim and J. Lee (2017)

This paper proposes a real-time object tracking method based on depth clustering. The method uses the depth information to segment the scene into different regions, and then groups the regions into distinct objects using a clustering algorithm. The method is evaluated on a dataset of videos captured by a Kinect sensor, and the results show that it can track objects in real-time with high accuracy.

4. "Object recognition and tracking using depth-based clustering and appearance-based modeling" by P. Singh, P. Kumar, and A. Jain (2016)

This paper proposes a method for object recognition and tracking using depth-based clustering and appearance-based modeling. The method first segments the scene into different regions based on the depth information, and then uses an appearance-based model to recognize and track the objects. The method is evaluated on a dataset of videos captured by a Kinect sensor, and the results show that it can accurately recognize and track objects in complex scenes.

5. "Object detection and tracking using depth-based clustering and Kalman filter" by S. Kim and J. Lee (2014)

This paper proposes a method for object detection and tracking using depth-based clustering and Kalman filter. The method first segments the scene into different regions based on the depth information, and then uses a Kalman filter to track the objects. The method is evaluated on a dataset of videos captured by a Kinect sensor, and the results show that it can accurately detect and track objects in complex scenes with high speed.