Benchmark data and method for real-time people counting in cluttered scenes using depth sensors.

Real-time automatic counting of people has widespread applications in security, safety and logistics.However, there is currently no large-scale public dataset for benchmarking approaches on this problem. To fill this gap, we introduce the first real-world RGB-D People Counting DataSet (PCDS) containing over 4,500 videos recorded at the entrance doors of buses in normal and cluttered conditions. We also propose an efficient method for counting people in real-world cluttered scenes using depth videos alone. The proposed method computes a point cloud from the depth video frame and re-projects it onto the ground plane to normalize the depth information. The resulting depth image is analyzed for identifying potential human heads. The human head proposals are meticulously refined using a 3D human model. The proposals in each frame of the continuous video stream are tracked to trace their trajectories. The trajectories are again refined to ascertain reliable counting. People are eventually counted by accumulating the head trajectories leaving the scene. To enable effective head and trajectory identification, we also propose two different compound features. Thorough evaluation on the PCDS demonstrates that our overall technique is able to count people in cluttered scenes with high accuracy at 45 fps on a 1.7 GHz processor.