Memorable and rich video summarization

Fei, Mengjuan; Jiang, Wei; Mao, Weijie

doi:10.1016/j.jvcir.2016.12.001

Cited by 50 publications

(25 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, an SVS approach is presented in [11] using clustering along with semantical, emotional, and shoot-quality clues for usergenerated summaries. Fei et al [12] used the fused score of memorability and entropy to generate final summary. Majority of the research in literature focus on SVS due to its simplicity as compared to MVS.…”

mentioning

confidence: 99%

Cloud-Assisted Multiview Video Summarization Using CNN and Bidirectional LSTM

Hussain

Muhammad

Ullah

et al. 2020

IEEE Trans. Ind. Inf.

120

View full text Add to dashboard Cite

The massive amount of video data produced by surveillance networks in industries instigate various challenges in exploring these videos for many applications such as video summarization (VS), analysis, indexing, and retrieval. The task of multi-view video summarization (MVS) is very challenging due to the gigantic size of data, redundancy, overlapping in views, light variations, and inter-view correlations. To address these challenges, various low-level features and clustering based soft computing techniques are proposed that cannot fully exploit MVS. In this article, we achieve MVS by integrating deep neural network based soft computing techniques in a two tier framework. The first online tier performs target appearance based shots segmentation and stores them in a lookup table that is transmitted to cloud for further processing. The second tier extracts deep features from each frame of a sequence in the lookup table and pass them to deep bi-directional long short-term memory (DB-LSTM) to acquire probabilities of informativeness and generate a summary. Experimental 1 evaluation on benchmark dataset and industrial surveillance data from YouTube confirms the better performance of our system compared to state-of-the-art MVS.

show abstract

mentioning

confidence: 99%

Cloud-Assisted Multiview Video Summarization Using CNN and Bidirectional LSTM

Hussain

Muhammad

Ullah

et al. 2020

IEEE Trans. Ind. Inf.

120

View full text Add to dashboard Cite

show abstract

“…Fei et al [21], the profound network predicted picture memorability score is presented and coupled with the keyframe extraction entropy value. Not only are the summaries generated semantically interesting, but they also maintain the videos' variety.…”

Section: Related Workmentioning

confidence: 99%

“…The transportation of the all video streaming is unrealistic due to the broader range (distance) between the base station and visual processing hub (VPH) due to bandwidth as well as the energy limitations of WMSNs. Researchers have used distinct compression [37], [38] and video summary methods [18]- [21] to address this problem in order to decrease the quantity of visual information at the visual processing hub (VPH) so that only most relevant video frames are furthered to the base station (BS) for processing. Considering energy constraints as well as the bandwidth, keyframe extraction employed to reduce data redundancy in which [16] expressed his thought in terms of salient motion detection.…”

Section: A Keyframe Extraction Model From Visual Datamentioning

confidence: 99%

See 1 more Smart Citation

SMSH: Secure Surveillance Mechanism on Smart Healthcare IoT System With Probabilistic Image Encryption

Khan

Ahamad

et al. 2020

IEEE Access

View full text Add to dashboard Cite

The Internet of Things is made of diverse networked things (i.e., smart, intelligent devices) that are consistently interconnected, producing meaningful data across the network without human interaction. Nowadays, the Healthcare system is widely interconnected with IoT environments to facilitate the best possible patient monitoring, efficient diagnosis, and timely operate with existing diseases towards the patients. Concerning the security and privacy of the patient's information. This paper is focused on Secure surveillance mechanism for a medical healthcare system with enabled internet of Things (sensors) by intelligently recorded video summary into the server and keyframes image encryption. This paper is twofold. Firstly, a well-organized keyframe extraction mechanism is called to extract meaningful image frames (detected normal/abnormal activities keyframe) by the visual sensor with an alert sent to the concerned authority in the healthcare system. Secondly, the final decision about the happened activity with extracted keyframes to keep highly secure from any adversary, and we propose an efficient probabilistic and lightweight encryption algorithm. Our proposed mechanism verifies effectiveness through producing results, robustness in nature, minimum execution time, and comparatively secure than other images (keyframes) encryption algorithms. Additionally, this mechanism can reduce storage, bandwidth, required transmission cost, and timely analysis of happened activity from any adversary with protecting the privacy of the patient's information in the IoT enabled healthcare system.

show abstract

“…The network is trained so that the reconstruction error is minimum. Fei et al in [12] proposed video summarization framework based on entropy and memorability score. The memorability score is computed using Hybrid-AlexNet.…”

Section: Introductionmentioning

confidence: 99%

Domain independent redundancy elimination based on flow vectors for static video summarization

Mohan

Nair

2019

Heliyon

View full text Add to dashboard Cite

Video summarization aims to find a compact representation of input videos. The method finds out interesting parts of the video by discarding the remaining parts of the video. The abstracts thus generated enhances browsing and retrieval of video data. The quality of summaries generated by video summarization algorithms can be improved if the redundant frames in the input video are taken care of before summarization. This paper presents a novel domain-independent method for redundancy elimination from input videos before summarization maintaining keyframes in the original video. The frames of input video are first presampled by selecting two frames in one second. The flow vectors between consecutive frames are computed using SIFT Flow algorithm. The magnitude of flow vectors at each pixel position of the frame are summed up to find the displacement magnitude between the consecutive frames. The redundant frames are filtered out based on local averaging of the displacement values. The evaluation of the method is done using two standard datasets namely VSUMM and OVP. The results demonstrate that an average reduction rate of 97.64% is achieved consistently on videos of all categories. The method also gives superior results compared to other state-of-the-art redundancy elimination methods for video summarization

show abstract

Memorable and rich video summarization

Cited by 50 publications

References 24 publications

Cloud-Assisted Multiview Video Summarization Using CNN and Bidirectional LSTM

Cloud-Assisted Multiview Video Summarization Using CNN and Bidirectional LSTM

SMSH: Secure Surveillance Mechanism on Smart Healthcare IoT System With Probabilistic Image Encryption

Domain independent redundancy elimination based on flow vectors for static video summarization

Contact Info

Product

Resources

About