Crowd Flow Segmentation in Compressed Domain using CRF
This work addresses crowd flow segmentation for video surveillance applications, offering an incremental improvement by applying an existing method (CRF) to compressed domain data.
The paper tackles the problem of segmenting crowd flows in H.264 compressed videos by proposing an unsupervised algorithm that models motion vectors as a Conditional Random Field (CRF) to obtain oriented motion segments, which are then merged based on gradient boundaries. The result demonstrates superior performance in accuracy and computational time on a standard dataset.
Crowd flow segmentation is an important step in many video surveillance tasks. In this work, we propose an algorithm for segmenting flows in H.264 compressed videos in a completely unsupervised manner. Our algorithm works on motion vectors which can be obtained by partially decoding the compressed video without extracting any additional features. Our approach is based on modelling the motion vector field as a Conditional Random Field (CRF) and obtaining oriented motion segments by finding the optimal labelling which minimises the global energy of CRF. These oriented motion segments are recursively merged based on gradient across their boundaries to obtain the final flow segments. This work in compressed domain can be easily extended to pixel domain by substituting motion vectors with motion based features like optical flow. The proposed algorithm is experimentally evaluated on a standard crowd flow dataset and its superior performance in both accuracy and computational time are demonstrated through quantitative results.