Video summarization using a key frame selection based on shot segmentation

Widiarto, Wisnu; Yuniarno, Eko Mulyanto; Hariadi, Mochamad

doi:10.1109/icsitech.2015.7407805

Cited by 5 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Segmentation-based algorithms primarily focus on the detection of substantial changes between successive frames. These methodologies, however, may face the issue of extracting repetitive key frames if recurring content is present in the video [24][25][26]. Where as clustering-based techniques involve grouping frames and selecting those nearest to the cluster centers as key frames.…”

Section: Related Workmentioning

confidence: 99%

Efficient visual information indexation for supporting actionable intelligence and knowledge generation

Nurye,

Aktas,

García

2023

Artificial Intelligence for Security and Defence Applications

View full text Add to dashboard Cite

Visual indexing, or the ability to search and analyze visual media such as images and videos, is important for law enforcement agencies because it can speed up criminal investigations. As more and more visual media is created and shared online, the ability to effectively search and analyze this data becomes increasingly important for investigators to do their job effectively. The major challenges for video captioning include accurately recognizing the objects and activities in the image, understanding their relationships and context, generating natural and descriptive language, and ensuring the captions are relevant and useful. Near real-time processing is also required in order to facilitate agile forensic decision making and prompt triage, hand-over and reduction of the amount of data to be processed by investigators or subsequent processing tools. This paper presents a captioning-driven efficient video analytic which is able to extract accurate descriptions of images and videos files. The proposed approach includes a temporal segmentation technique providing the most relevant frames. Subsequently, an image captioning approach has been specialized to describe visual media related to counter-terrorism and cybercrime for each relevant frame. Our proposed method achieves high consistency and correlation with human summary on SumMe dataset, outperforming previous similar methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Efficient visual information indexation for supporting actionable intelligence and knowledge generation

Nurye,

Aktas,

García

2023

Artificial Intelligence for Security and Defence Applications

View full text Add to dashboard Cite

show abstract

“…While these shortcomings were taken into account, it was noticed that the majority of available literature does not expand on the sensitivity of the parameter values they employ. One such an example is the work presented by Widiarto et al in [28] where they implemented a histogram based approach for video segmentation by calculating histograms for each of the RGB (Red-Green-Blue) color components of the video. There is however no information, nor justification provided as to the chosen threshold used in conjunction with the Euclidean distance metric which is calculated from the histograms.…”

Section: Literature Overviewmentioning

confidence: 99%

“…There is however no information, nor justification provided as to the chosen threshold used in conjunction with the Euclidean distance metric which is calculated from the histograms. While the work presented in [28] utilises a video segmentation algorithm, the focus seems to be on subsequent processing of the video, with no useful information pertaining to the video segmentation algorithm.…”

Section: Literature Overviewmentioning

confidence: 99%

Parameter Analysis of the Jensen-Shannon Divergence for Shot Boundary Detection in Streaming Media Applications

Klerk¹,

Venter²,

Hoffman³

2018

SAIEE Afr. Res. J.

View full text Add to dashboard Cite

Shot boundary detection is an integral part of multimedia, be it video management or video processing. Multiple boundary detection techniques have been developed throughout the years, but are only applicable to very specific instances. The Jensen-Shannon divergence (JSD) is one such a technique that can be implemented to detect the shot boundaries in digital videos. This paper investigates the use of the JSD algorithm to detect shot boundaries in streaming media applications. Furthermore, the effects of the various parameters used by the JSD technique, on the accuracy of the detected boundaries, are quantified by the recall and precision metrics all the while keeping track of how they affect the execution time.

show abstract

“…Shot can be detected using transitions between successive frames in a video. Shot detection techniques have been done based on the color histogram [18], based on pixel differences [19], and motion information [20].…”

Section: Introductionmentioning

confidence: 99%

Keyframe Selection of Frame Similarity to Generate Scene Segmentation Based on Point Operation

Widiarto¹,

Hariadi²,

Yuniarno³

2018

IJECE

Self Cite

View full text Add to dashboard Cite

<span lang="EN-US">Video segmentation has been done by grouping similar frames according to the threshold. Two-frame similarity calculations have been performed based on several operations on the frame: point operation, spatial operation, geometric operation and arithmatic operation. In this research, similarity calculations have been applied using point operation: frame difference, gamma correction and peak signal to noise ratio. Three-point operation has been performed in accordance with the intensity and pixel frame values. Frame differences have been operated based on the pixel value level. Gamma correction has analyzed pixel values and lighting values. The peak signal to noise ratio (PSNR) has been related to the difference value (noise) between the original frame and the next frame. If the distance difference between the two frames was smaller then the two frames were more similar. If two frames had a higher gamma correction factor, then the correction factor would have an increasingly similar effect on the two frames. If the value of PSNR was greater then the comparison of two frames would be more similar. The combination of the three point operation methods would be able to determine several similar frames incorporated in the same segment</span>

show abstract

Video summarization using a key frame selection based on shot segmentation

Cited by 5 publications

References 17 publications

Efficient visual information indexation for supporting actionable intelligence and knowledge generation

Efficient visual information indexation for supporting actionable intelligence and knowledge generation

Parameter Analysis of the Jensen-Shannon Divergence for Shot Boundary Detection in Streaming Media Applications

Keyframe Selection of Frame Similarity to Generate Scene Segmentation Based on Point Operation

Contact Info

Product

Resources

About