Crowd flow segmentation based on motion vectors in H.264 compressed domain

Praveen, R. Gnana; Babu, R. Venkatesh

doi:10.1109/conecct.2014.6740330

Cited by 5 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence we propose a novel framework of achieving flow segmentation in the compressed domain using motion vectors. The prospect of achieving crowd flow segmentation in the compressed domain was explored in [14]. However, on the pursuit of improving the computational efficiency, a novel framework with less computational complexity still retaining comparable accuracy was proposed and demonstrated on the dataset provided by [1].…”

Section: Related Workmentioning

confidence: 99%

Super-pixel based crowd flow segmentation in H.264 compressed videos

Biswas

Praveen

Babu

2014

2014 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

In this paper, we have proposed a simple yet robust novel approach for segmentation of high density crowd flows based on super-pixels in H.264 compressed videos. The collective representation of the motion vectors of the compressed video sequence is transformed to color map and super-pixel segmentation is performed at various scales for clustering the coherent motion vectors. The number of dynamically meaningful flow segments is determined by measuring the confidence score of the accumulated multi-scale super-pixel boundaries. The final crowd flow segmentation is obtained from the edges that are consistent across all the super-pixel resolutions. Hence, our major contribution involves obtaining the flow segmentation by clustering the motion vectors and determination of number of flow segments using only motion super-pixels without any prior assumption of the number of flow segments. The proposed approach was benchmarked on standard crowd flow dataset. Experiments demonstrated better accuracy and speedup for the proposed approach compared to the state-of-the-art methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Super-pixel based crowd flow segmentation in H.264 compressed videos

Biswas

Praveen

Babu

2014

2014 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Ali et al [2] proposed an approach, in pixel-domain, to segment the flows by observing the particle flows over optical flow field. Praveen et al [26] tackle this problem in compressed domain by clustering the dominant pattern of motion vectors using Expectation-Maximization algorithm. The clusters which converge to a single flow are merged together based on the Bhattacharya distance measure between the histogram of the orientation of the motion vectors at the boundaries of the clusters.…”

Section: Crowd Flow Segmentationmentioning

confidence: 99%

A survey on compressed domain video analysis techniques

Babu

Tom

Wadekar

2014

Multimed Tools Appl

Self Cite

View full text Add to dashboard Cite

Image and video analysis requires rich features that can characterize various aspects of visual information. These rich features are typically extracted from the pixel values of the images and videos, which require huge amount of computation and seldom useful for real-time analysis. On the contrary, the compressed domain analysis offers relevant information pertaining to the visual content in the form of transform coefficients, motion vectors, quantization steps, coded block patterns with minimal computational burden. The quantum of work done in compressed domain is relatively much less compared to pixel domain. This paper aims to survey various video analysis efforts published during the last decade across the spectrum of video compression standards. In this survey, we have included only the analysis part, excluding the processing aspect of compressed domain. This analysis spans through various computer vision applications such as moving object segmentation, human action recognition, indexing, retrieval, face detection, video classification and object tracking in compressed videos.

show abstract

“…To segment flows, an approach conceived by Ali et al [41] in pixel-domain. Praveen et al [42] handled this problem via invoking an EM algorithm. In a collective representation of compressed video series, research contributions made by the authors [43], to attain the stream division.…”

Section: Background Studies On Crowd Flow Segmentation / Dominant Flomentioning

confidence: 99%

Research Inferences In Video Compression Domain

Nihal*¹

2019

IJEAT

View full text Add to dashboard Cite

Video Compression (VC) is a painstaking investigation area in the modernization of digital existence because of more evolution being emerged in web relevance. In this direction, video compression is grown-up rapdily and confirmed by the large number of myriad applications in Video streaming, Computer Vision like Video Tampering Detection, Video Surveillance and Camera Moving etc., which collectively use of this compression technology. This paper attributes / opensup / endeavours the whole possible and recent research directions and its challenges by focusing the potential investigations on compression domain of MPEG, H.264, H.265 compression coding standard with latest computing techniques in learning videos via Machine Learning.

show abstract

Crowd flow segmentation based on motion vectors in H.264 compressed domain

Cited by 5 publications

References 9 publications

Super-pixel based crowd flow segmentation in H.264 compressed videos

Super-pixel based crowd flow segmentation in H.264 compressed videos

A survey on compressed domain video analysis techniques

Research Inferences In Video Compression Domain

Contact Info

Product

Resources

About