Compressed-domain video parsing using energy histograms of the lower-frequency DCT coefficients

1999 ◽  
Author(s):  
Oliver K. Bao ◽  
Jose A. Lay ◽  
Ling Guan
Keyword(s):  
Compressed Domain ◽  
Video Parsing ◽  
Dct Coefficients

The Study of Effective Video Summarization Algorithm

2012 ◽  
Vol 263-266 ◽  
pp. 2364-2368
Author(s):  
Dong Lin Ma ◽  
Xi Jun Zhang ◽  
Qian Mi
Keyword(s):  
Information System ◽  
Video Analysis ◽  
Video Summarization ◽  
Image Sequences ◽  
Video Sequences ◽  
Video Frame ◽  
Compressed Domain ◽  
The Core ◽  
Video Frames ◽  
Dct Coefficients

In this paper, a video summarization representation algorithm was proposed in compressed domain. In particular, Rough sets(RS) theory is introduced for video analysis to increase. Firstly, DCT coefficients and DC coefficients are extracted from video image sequences, so an Information System can construct with DC coefficients. Then Information System is reduced by ruduction theory of RS, the representation of the video frame is obtained by reduced DC coefficients. Finally, we can obtain the reduced Information System, i.e. the Core of Information System. Since the Core contained all the information in video sequences, and at the same time it banished redundant video frame, so it can be viewed as the effective summarization representation. Experimental results indicate that the algorithm can efficiently generate a set of summarization representative of videos sequences and enjoys following advantages. Only a subset of video frames considered during video analysis, so it can avoid the computational complexity, the video summarization representation becomes more scientific than previous methods.

Efficient Scene Change Algorithm Based on Statistical Feature in the DCT Domain

2011 ◽  
Vol 204-210 ◽  
pp. 229-233
Author(s):  
Jian Feng Wang ◽  
Jian Min Jiang
Keyword(s):  
Processing Speed ◽  
Abrupt Change ◽  
Feature Vector ◽  
Detection Algorithm ◽  
Compressed Domain ◽  
Scene Change ◽  
Scene Change Detection ◽  
Statistical Feature ◽  
Dct Domain ◽  
Dct Coefficients

In this paper, we propose an effective scene change detection algorithm directly in compressed domain. The proposed scene change algorithm(including abrupt change and ) test the n frame and n+1 frame through the extracting the feature of each frame. When extract the features of frame, two-dimensional statistical feature m1-s from DCT coefficients without its inverse transform was computed, Divide m1-s space into 42 unequal partitions (subspaces) and count the numbers pr within the 42 subspaces (entries) as the feature vector to judge the different of the two frames. locating scene changes is operated by comparison tests. In comparison with existing representative techniques, the experimental results show the superiority of the proposed method in terms of precision and processing speed.

scholarly journals Classification of Compressed Domain Images Utilizing Open VINO Inference Engine

2019 ◽  
Vol 9 (1) ◽  
pp. 1669-1678
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Computer System ◽  
Inference Engine ◽  
Intermediate Representation ◽  
Compressed Domain ◽  
Jpeg Images ◽  
Testing Accuracy ◽  
Dct Coefficients

This paper provides a platform to investigate and explore method of ‘partial decoding of JPEG images’ for image classification using Convolutional Neural Network (CNN). The inference is targeting to run on computer system with x86 CPU architecture. We aimed to improve the inference speed of classification by just using part of the compressed domain image information for prediction. We will extract and use the ‘Discrete Cosine Transform’ (DCT) coefficients from compressed domain images to train our models. The trained models are then converted into OpenVINO Intermediate Representation (IR) format for optimization. During inference stage, full decoding is not required as our model only need DCT coefficients which are presented in the process of image partial decoding. Our customized DCT model are able to achieve up to 90% validation and testing accuracy with great competence towards the conventional RGB model. We can also obtain up to 2x times inference speed boost while performing inference on CPU in compressed domain compared with spatial domain employing OpenVINO inference engine.

A Novel Hierarchical Dynamic Video Summarization Representation for Video Analysis

2012 ◽  
Vol 490-495 ◽  
pp. 465-469
Author(s):  
Xiang Wei Li ◽  
Yu Xiu Kang ◽  
Gang Zheng
Keyword(s):  
Information System ◽  
Video Analysis ◽  
Video Summarization ◽  
Image Sequences ◽  
Video Frame ◽  
Original Video ◽  
Compressed Domain ◽  
Video Information ◽  
Video Frames ◽  
Dct Coefficients

Based on Rough Sets (RS), a novel effective video summarization representation was proposed for video analysis in compressed domain. Firstly, DCT coefficients and DC coefficients are extracted from original video image sequences, so an Information System can construct with DC coefficients. Then, Information System is reduced by attributes reduction theory of RS, the representation of the video frame is achieved by reduced DC coefficients. Finally, the reduced Information System can be achieved. Since the Core of Information System contained all major video information in video sequences, which banished the redundant video frame, so it can be considered as the efficient summarization representation. Compared to conventional or existing algorithm, the algorithm enjoys following advantages. (1) Only a subset of video frames considered during video analysis, so it can avoid the computational complexity. (2) The video summarization representation becomes more scientific and efficient than previous methods. (3) According to the reduced frame number, the algorithm can extract hierarchical dynamic video summarization representation.

DENSE ESTIMATION OF OPTICAL FLOW FIELD WITHIN THE MPEG-2 COMPRESSED DOMAIN

2009 ◽  
Vol 09 (03) ◽  
pp. 435-448
Author(s):  
GAOBO YANG ◽  
WEIWEI CHEN ◽  
XIAO JING WANG ◽  
ZHAOYANG ZHANG
Keyword(s):  
Flow Field ◽  
Optical Flow ◽  
Motion Vector ◽  
Compressed Domain ◽  
Motion Vectors ◽  
Optical Flow Field ◽  
Intensity Information ◽  
Motion Vector Field ◽  
First Motion ◽  
Dct Coefficients

A dense estimation of optical flow field within the MPEG-2 compressed domain is proposed, which utilizes only the compressed-domain information, i.e. motion vectors and DCT coefficients. First, motion vectors are pre-processed to estimate the DCT coefficients for P and B frames. Second, initial optical flow is estimated with Black's optical flow estimation framework, in which DC image is substituted by DC+2AC image to provide more intensity information. Third, high confidence test is exploited to generate the dense and accurate motion vector field by removing false and noisy motion vectors. It preserves the advantages of compressed domain processing and improves the existent MPEG velocity field in terms of accuracy and density. Experimental results demonstrate that the proposed approach can provide a satisfactory motion analysis for compressed-domain video object extraction.

Sign in / Sign up

Export Citation Format

Share Document