Content-based Copy Detection Research Articles

Video fingerprinting: Past, present, and future

The last decades have seen video production and consumption rise significantly: TV/cinematography, social networking, digital marketing, and video surveillance incrementally and cumulatively turned video content into the predilection type of data to be exchanged, stored, and processed. Belonging to video processing realm, video fingerprinting (also referred to as content-based copy detection or near duplicate detection) regroups research efforts devoted to identifying duplicated and/or replicated versions of a given video sequence (query) in a reference video dataset. The present paper reports on a state-of-the-art study on the past and present of video fingerprinting, while attempting to identify trends for its development. First, the conceptual basis and evaluation frameworks are set. This way, the methodological approaches (situated at the cross-roads of image processing, machine learning, and neural networks) can be structured and discussed. Finally, fingerprinting is confronted to the challenges raised by the emerging video applications (e.g., unmanned vehicles or fake news) and to the constraints they set in terms of content traceability and computational complexity. The relationship with other technologies for content tracking (e.g., DLT - Distributed Ledger Technologies) are also presented and discussed.

A Study of the Possibility of Usage Motion Vectors of Compressed Videos to Create Video Identification

The use of motion vectors for identifying video sequences has been well studied (in the framework of research on the topic CBCD – Content-Based Copy Detection ‒ detecting copies of videos based on content analysis). This makes it possible to check the similarity of two video fragments or search for a fragment in a larger video sequence. Existing and well-known methods for forming identification datasets typically use complete video stream decoding. The authors suggested using the motion vectors of a compressed video stream, which reduces the computational costs for identifying video sequences and uses simplified algorithms to generate identification data. Unlike the previously proposed methods, which implement either modified video codecs or obsolete ones, the authors propose using data formed by compression codecs that are used in the most common video hosting platforms (Youtube, Vimeo, etc.) The possibility of forming an automated system of comparing video sequences, along with its possibilities and limitations, will be studied in the following works.

Attack Detection on Images Based on DCT-Based Features

As reproduction of images can be done with ease, copy detection has increasingly become important. In the duplication process, image modifications are likely to occur and some alterations are deliberate and can be viewed as attacks. A wide range of copy detection techniques has been proposed. In our study, content-based copy detection, which basically applies DCT-based features for images, namely, pixel values, edges, texture information and frequency-domain component distribution, is employed. Experiments are carried out to evaluate robustness and sensitivity of DCT-based features from attacks. As different types of DCT-based features hold different pieces of information, how features and attacks are related can be shown in their robustness and sensitivity. Rather than searching for proper features, use of robustness and sensitivity is proposed here to realize how the attacked features have changed when an image attack occurs. The experiments show that, out of ten attacks, the neural networks are able to detect seven attacks namely, Gaussian noise, S&P noise, Gamma correction (high), blurring, resizing (big), compression and rotation with mostly related to their sensitive features.

CNN Feature-Based Image Copy Detection with Contextual Hash Embedding

As one of the important techniques for protecting the copyrights of digital images, content-based image copy detection has attracted a lot of attention in the past few decades. The traditional content-based copy detection methods usually extract local hand-crafted features and then quantize these features to visual words by the bag-of-visual-words (BOW) model to build an inverted index file for rapid image matching. Recently, deep learning features, such as the features derived from convolutional neural networks (CNN), have been proven to outperform the hand-crafted features in many applications of computer vision. However, it is not feasible to directly apply the existing global CNN features for copy detection, since they are usually sensitive to partial content-discarded attacks, such as copping and occlusion. Thus, we propose a local CNN feature-based image copy detection method with contextual hash embedding. We first extract the local CNN features from images and then quantize them to visual words to construct an index file. Then, as the BOW quantization process decreases the discriminability of these features to some extent, a contextual hash sequence is captured from a relatively large region surrounding each CNN feature and then is embedded into the index file to improve the feature’s discriminability. Extensive experimental results demonstrate that the proposed method achieves a superior performance compared to the related works in the copy detection task.

Multimedia Signal Processing in Cloud using Content Based Copy Detection

Nowadays making multimedia and multimedia altering has gotten simple because of effectively accessible handling instruments. Progressively over facilitating sites are unreservedly accessible. So clearly we experience circumstances wherein numerous recordings and pictures that are copyrighted are getting copied effectively. Such duplication isn't just unlawful yet in addition it makes an enormous misfortune unique content holder. On the off chance that at all we attempt to distinguish such copied content, process gets extremely mind boggling in light of enormous sizes and huge amount of multimedia documents present on web. So this procedure gets perplexing and computationally costly. Content based multimedia copy detection gives a powerful and improved instrument which can distinguish coordinating rate among tremendous multimedia duplicates from the content itself without having any reliance on any product. In the system of content based copy detection, marks are made from content itself before transferring on cloud like Picasa. In the proposed framework at whatever point new Multimedia article is transferred on the cloud initially its mark is made and on the off chance that the mark matches with put away database signature, at that point this will trap the site which copies copyrighted copy with no rights.

Shearlet Based Video Fingerprint for Content-Based Copy Detection

Content-based copy detection (CBCD) is widely used in copyright control for protecting unauthorized use of digital video and its key issue is to extract robust fingerprint against different attacked versions of the same video. In this paper, the “natural parts” (coarse scales) of the Shearlet coefficients are used to generate robust video fingerprints for content-based video copy detection applications. The proposed Shearlet-based video fingerprint (SBVF) is constructed by the Shearlet coefficients in Scale 1 (lowest coarse scale) for revealing the spatial features and Scale 2 (second lowest coarse scale) for revealing the directional features. To achieve spatiotemporal natural, the proposed SBVF is applied to Temporal Informative Representative Image (TIRI) of the video sequences for final fingerprints generation. A TIRI-SBVF based CBCD system is constructed with use of Invert Index File (IIF) hash searching approach for performance evaluation and comparison using TRECVID 2010 dataset. Common attacks are imposed in the queries such as luminance attacks (luminance change, salt and pepper noise, Gaussian noise, text insertion); geometry attacks (letter box and rotation); and temporal attacks (dropping frame, time shifting). The experimental results demonstrate that the proposed TIRI-SBVF fingerprinting algorithm is robust on CBCD applications on most of the attacks. It can achieve an average F1 score of about 0.99, less than 0.01% of false positive rate (FPR) and 97% accuracy of localization.

A spectrogram-based audio fingerprinting system for content-based copy detection

This paper presents a novel audio fingerprinting method that is highly robust to a variety of audio distortions. It is based on an unconventional audio fingerprint generation scheme. The robustness is achieved by generating different versions of the spectrogram matrix of the audio signal by using a threshold based on the average of the spectral values to prune this matrix. We transform each version of this pruned spectrogram matrix into a 2-D binary image. Multiple versions of these 2-D images suppress noise to a varying degree. This varying degree of noise suppression improves likelihood of one of the images matching a reference image. To speed up matching, we convert each image into an n-dimensional vector, and perform a nearest neighbor search based on this n-dimensional vector. We give results with two different feature parameters and their combination. We test this method on TRECVID 2010 content-based copy detection evaluation dataset, and we validate the performance on TRECVID 2009 dataset also. Experimental results show the effectiveness of these features even when the audio is distorted. We compare the proposed method to two state-of-the-art audio copy detection systems, namely NN-based and Shazam systems. Our method by far outperforms Shazam system for all audio transformations (or distortions) in terms of detection performance, number of missed queries and localization accuracy. Compared to NN-based system, our approach reduces minimal Normalized Detection Cost Rate (min NDCR) by 23 % and improves localization accuracy by 24 %.

Fast beta wavelet network-based feature extraction for image copy detection

The applications of authors' rights protection against illegal generation of images do not cease to evolve. However, most of them impose a very high computational cost especially when working with a database containing thousands of images. This paper addresses the problem of authors' rights violation and presents an original scheme, for Content-Based Image Copy Detection (CBICD), based on two screens: approximation screen and details screen. These two screens, based on Fast Beta Wavelet Transform (FBWT), aimed to filter the original images based first on their approximation and then on detail appearances respectively to display the corresponding original image to a given query one (a copy image). Extensive experiments of 8118 images from Copydays and Holidays datasets generated by INREA11http://www.lear.inrialpes.fr/jegou/data.php proved the effectiveness as well as the search speed of our approach in CBICD.

Content-Based Video Copy Detection Benchmarking at TRECVID

This article presents an overview of the video copy detection benchmark which was run over a period of 4 years (2008--2011) as part of the TREC Video Retrieval (TRECVID) workshop series. The main contributions of the article include i) an examination of the evolving design of the evaluation framework and its components (system tasks, data, measures); ii) a high-level overview of results and best-performing approaches; and iii) a discussion of lessons learned over the four years. The content-based copy detection (CCD) benchmark worked with a large collection of synthetic queries, which is atypical for TRECVID, as was the use of a normalized detection cost framework. These particular evaluation design choices are motivated and appraised.

Content-based video copy detection based on motion vectors estimated using a lower frame rate

We propose a motion vector-based video content-based copy detection method. One of the signatures of a given video is motion vectors extracted from image sequences. However, when consecutive image frames are used, the resulting motion vectors are not descriptive enough because most vectors are either too small or they appear to scatter in all directions. We calculate motion vectors in a lower frame rate than the actual frame rate of the video to overcome this problem. As a result, we obtain large vectors and they represent a given video in a robust manner. We carry out experiments for various parameters and present the results.

The System to Retain Procurement of Video Data and Eradicate UN trusted User’s

In this paper a secure mechanism to protect the intellectual property of the content creator of the websites like Google video, you tube, yahoo. Here the server will identify the unauthorized user and block that user from the network. By using the video copy detection methods compare the user video and admin videos to detect whether the trained video and queried videos may copies are not. Initially water marking techniques are used to protect the intellectuality of admin, but water marking itself is fragile for compression. Then content based copy detection (CBCD) proposed, it has a signature limitation problem. So the local descriptor Scale Invariant Feature Transform (SIFT) introduce to extort the features from the key frames. It encounters a computational cost. Then Singular Value Decomposition (SVD) is proposed for match two frames of video by using SIFT descriptor. Auto dual brink process is used to eradicate the superfluous video frames. And the graph based process is used to obtain video sequence and identify the longest path of matching result along with timing constraint. Then the queried video is the copy then that particular user blocked based on time stamp Keywords: Blocking, Intellectual property, video copy, Superfluous frames, Time Stamp.

A Survey Of Sign Based Image Copy Detection Methods

The world wide web is filled with billions of images and redundant copies of images can frequently be found on many websites. These duplicates can be exact copies or differ slightly in their visual content. To assure privacy it is mandatory to preserve copyright verification, image-content identification, copy detection and authentication. This paper provides a comparative study on how content-based image copy detection methods are used to detect the copy of original image. The methods based on content based copy detection which includes signs of discrete cosine transform (DCT) and discrete wavelet transform (DWT), color histograms, by using concept of dual signature are presented.

Content-based copy detection through multimodal feature representation and temporal pyramid matching

Content-based copy detection (CBCD) is drawing increasing attention as an alternative technology to watermarking for video identification and copyright protection. In this article, we present a comprehensive method to detect copies that are subjected to complicated transformations. A multimodal feature representation scheme is designed to exploit the complementarity of audio features, global and local visual features so that optimal overall robustness to a wide range of complicated modifications can be achieved. Meanwhile, a temporal pyramid matching algorithm is proposed to assemble frame-level similarity search results into sequence-level matching results through similarity evaluation over multiple temporal granularities. Additionally, inverted indexing and locality sensitive hashing (LSH) are also adopted to speed up similarity search. Experimental results over benchmarking datasets of TRECVID 2010 and 2009 demonstrate that the proposed method outperforms other methods for most transformations in terms of copy detection accuracy. The evaluation results also suggest that our method can achieve competitive copy localization preciseness.

Content-based Plagiarism Detection in Korean Document Using Ferret��s Trigram

Document plagiarism means the unauthorized use of the original document of another author without recognition of the source. With the development of the Internet, the volume of digital information available and easily accessible has increased massively and detecting plagiarism manually is so expensive in terms of both time and effort. Although many copy detection techniques for digital document already have been released, their performance is still unsatisfactory. This paper proposes content-based copy detection for Hangul (Korean character) documents by improving the detection accuracy of existing Ferret’s trigram. The key of the proposed system to identify plagiarism is to use two elements: firstly the number of matching trigrams in the original document and secondly the weighting factor of the trigrams where they match sequentially. In this study we show that the proposed system is developed further by weighting results depending on the matching degree of trigram, thereby improving the accuracy of similarity detection in Hangul documents.

Spider

This article presents a novel content-based copy detection system for 3D videos. The system creates compact and robust depth and visual signatures from the 3D videos. Then, signature of a query video is compared against an indexed database of reference videos' signatures. The system returns a score, using both spatial and temporal characteristics of videos, indicating whether the query video matches any video in the reference video database, and in case of matching, which portion of the reference video matches the query video. Analysis shows that the system is efficient, both computationally and storage-wise. The system can be used, for example, by video content owners, video hosting sites, and third-party companies to find illegally copied 3D videos. We implemented Spider, a complete realization of the proposed system, and conducted rigorous experiments on it. Our experimental results show that the proposed system can achieve high accuracy in terms of precision and recall even if the 3D videos are subjected to several transformations at the same time. For example, the proposed system yields 100% precision and recall when copied videos are parts of original videos, and more than 90% precision and recall when copied videos are subjected to different individual transformations.

Content-based copy detection by a subspace learning based video fingerprinting scheme

We propose a video copy detection scheme that employs a transform domain global video fingerprinting method. Video fingerprinting has been performed by the subspace learning based on nonnegative matrix factorization (NMF). It is shown that the binary video fingerprints extracted from the basis and gain matrices of the NMF representation enable us to efficiently represent the spatial and temporal content of a video segment respectively. An extensive performance evaluation has been carried out on the query and reference dataset of CBCD task of TRECVID 2011. Our results are compared with the average and the best performance reported for the task. Also NDCR and F1 rates are reported in comparison to the performance achieved via the global methods designed by the TRECVID 2011 participants. Results demonstrate that the proposed method achieves higher correct detection rates with good localization capability for the transformation of text/logo insertion, strong re-encoding, frame dropping, noise addition, gamma change or their mixtures; however there is still potential for improvement to detect copies with picture-in-picture transformations. It is also concluded that the introduced binary fingerprinting scheme is superior to the existing transform based methods in terms of the compactness.

A Novel CBCD Method Based on 3-Dimensional SIFT Descriptors in Dynamic Tunnels

Content-Based Copy Detection (CBCD) is a heated research area. A variety of features are extracted and applied extensively to improve the performance of CBCD systems. A novel scheme which adopts 3-Dimensional SIFT Descriptors (Paul Scavenger, 2007) based on the innovative dynamic tunnels is proposed in this paper. The algorithm is aim to reach a proper balance between efficiency and accuracy. Experimental results indicate that the algorithm achieves an accuracy match with inexpensive computational cost.

Efficient Spatiotemporal Matching for Video Copy Detection in H.264/AVC Video

for the H.264/AVC standard. The mechanism is based on content based copy detection (CBCD). The proposed method divides each frame within a group of three consecutive frames into a grid. Each corresponding grid across these groups of frames is then sorted in an ordinal vector which describes both, the spatial as well as the temporal variation. This ordinal matrix based copy-detection scheme is effective in detecting not only a copied video clip but also its location within a longer video sequence. The technique has been designed to work in the compressed domain which makes it computationally very efficient. The proposed mechanism was tested on a number of video sequences containing copies which had undergone a variety of modifications. The results proved that the proposed technique is capable of detecting these copies effectively and efficiently and hence is suitable for forensic applications.

A Study on Video Copy Detection Methods Using Representative Color Sequence for Protecting Copyrights

This study deals with the technical approach related to the protection of digital video contents copyright, which has currently become a major issue in the digital contents market. With the production and distribution of digital video contents rapidly increasing, various policies and technical approaches are being made in relation to the protection of intellectual property rights. The technical approach from a DRM (Digital Rights Management) perspective was made from various angles, but its limitations have also been exposed in terms of convenience and quality. In this study, rather than the existing encoding certification and watermark technologies, it introduces technical approaches from a content-based copy detection method, while suggesting a copy detection method using the representative color sequence. In particular, this study takes into consideration the distribution environment of the domestic video market and deals with the technical approach in a direction that can control the illegal distribution of high quality videos.

Audio-visual large-scale video copy detection

With the development of hardware and proliferation of Web 2.0, an increasing need to provide efficient access of desired content in large-scale database has emerged. Content-based copy detection (CBCD) has attracted more research interests due to its broad applications. However, most existing CBCD approaches only focus on single view (visual view or audio view). In this paper, we attempt to address the CBCD task by considering both visual and audio information simultaneously. A bag of visual concept words model for video copy detection method is proposed from visual cue, and the main advantage of the proposed method lies in the fact that the presence of semantic concepts is highly robust to spatial and temporal video transformations. On the other hand, a bag of audio words model is presented from audio cue and a coherency vocabulary combined with soft-weighted strategy realizes fast and accurate indexing. Finally, the late fusion is adopted to obtain the final audio–visual result based on visual-only and audio-only results. Intensive experiments are conducted on two large-scale data sets and competitive results are achieved.