Film grain, once a by-product of analog film, is now present in most cinematographic content for aesthetic reasons. However, when such content is compressed at medium to low bitrates, film grain is lost due to its random nature. To preserve artistic intent while compressing efficiently, film grain is analyzed and modeled before encoding and synthesized after decoding. This paper introduces FGA-NN, the first learning-based film grain analysis method to estimate conventional film grain parameters compatible with conventional synthesis. Quantitative and qualitative results demonstrate FGA-NN’s superior balance between analysis accuracy and synthesis complexity, along with its robustness and applicability.

End-to-end image and video compression using auto-encoders (AE) offers new appealing perspectives in terms of rate-distortion gains and applications. While most complex models are on par with the latest compression standard like VVC/H.266 on objective metrics, practical implementation and complexity remain strong issues for real-world applications. We propose a practical implementation suitable for realistic applications. We demonstrate that some gains can be achieved on top low-complexity AE, even when using simpler implementation. The proposed implementation also allows a direct integration of such approaches on a variety of platforms and code is made available as a pure C++ standalone codec [1]:

In this work we extend the idea of object co-segmentation [10] to perform interactive video segmentation. Our framework predicts the coordinates of vertices along the boundary of an object for two frames of a video simultaneously. The predicted vertices are interactive in nature and a user interaction on one frame assists the network to correct the predictions for both frames. We employ attention mechanism at the encoder stage and a simple combination network at the decoder stage which allows the network to perform this simultaneous correction efficiently. The framework is also robust to the distance between the two input frames as it can handle a distance of up to 50 frames in between the two inputs.We train our model on professional dataset, which consists pixel accurate annotations given by professional Roto artists. We test our model on DAVIS [15] and achieve state of the art results in both automatic and interactive mode surpassing Curve-GCN [11] and PolyRNN++ [1].

Designing watermarks that can withstand geometric and temporal attacks is a notoriously difficult problem. The registration of a pirate content with the reference content, prior to the decoding of an embedded forensic mark, is therefore preferred. Semi-automatic registration systems, requiring the expert user to manually select and match control points through a suitable user interface, are not acceptable for content owners and do not usually provide very accurate results. A fully automated registration scheme is therefore chosen to synchronize pirate and master contents. By extracting robust and unique features, digital fingerprinting provides a relevant framework for automatic semi-blind content registration. In this paper we focus on temporal registration. In order to guarantee an accurate alignment while keeping the fingerprint size as small as possible, we propose a temporally adaptive approach. The registration scheme is designed around an adaptive motion description based on hierarchical encoding of the wavelet coefficients computed on the difference of successive frames.

In this industrial exhibit, Technicolor will showcase various technologies developed to protect digital audio-visual material all along the media value chain, from production to distribution, including forensics services in case pirated content is detected. The exhibit will be organized around Technicolor's proprietary technologies in video fingerprinting and audio/video watermarking.

This paper presents a framework for the forensic analysis of video content, for example those illegal copies of feature films proliferating in peer-to-peer networks or on bootleg DVDs. Compact and discriminant feature vectors, so called ¿fingerprints¿, are computed for a catalog of master contents. They combine global (image-based) and local (keypoint-based) signal properties extracted on representative keyframes, together with a one-dimensional signature capturing the temporal evolution of the video. We demonstrate that the extracted fingerprints not only allow accurate copy detection but also enable precise temporal and geometric alignment of candidate videos with master content. Besides making possible watermark decoding in the presence of geometric distortion, this opens the door to the reconstruction of the history of signal manipulations. The performance of the proposed scheme is assessed through a realistic copy detection and registration scenario.

Selective encryption is a new trend in content protection. It aims at reducing the amount of data to encrypt while achieving a sufficient and inexpensive security. This approach is particularly desirable in constrained communication (real time networking with delay constraints, mobile communication with limited computational power...). In this paper we introduce selective encryption from information theory point of view. We define a set of evaluation criteria for selective encryption algorithms and propose a novel selective encryption algorithm for JPEG2000 compressed images satisfying all these criteria. The main contribution of this proposal consists of reaching the minimum amount of data to encrypt regarding a given level of security and target application requirements. For this purpose, we exploit the R-D optimization performed by JPEG2000 EBCOT algorithm.

In this paper, we study the security of an image scrambling algorithm based on space-filling curves (SFC). A random SFC is a pixel permutation that changes the scanning order without changing pixels values. Few attacks were reported in the literature. A ciphertext-only attack when a different scan is generated for each frame in the video was proposed. The success of this attack heavily depends on the statistics of the plain frame. In addition, scanning each frame with a different SFC is not adapted to compression as many compression algorithms exploit temporal redundancy. We revisit the security properties of this scrambling technique and propose an efficient and low-cost chosen-plaintext attack when the same SFC is used for each frame in the video.

The proliferation of pirate copies of feature films on peer-to-peer networks arouses a great interest to countermeasures such as the insertion of (invisible) forensic marks in projected movies, to deter their illegal capture. The registration of pirate copies with the original content is however a prerequisite to the recovery of such embedded messages, as severe geometric distortions often occur in illegally camcorded contents. After a brief state-of-the-art in image registration, the paper details an algorithm for video registration, focusing on the compensation of geometric distortions. Control points are automatically extracted in original and copy pictures, followed by pre- and post-matching filtering steps to discard not relevant control points and erroneous matched pairs of control points respectively. This enables the accurate numerical estimation of an 8-parameter homographic distortion model, used to register the copy frames with the original reference grid. Such an image registration algorithm is inserted into a general video registration scheme. Results are presented on both natural and synthetic test material.

A fingerprinting design extracts discriminating features, called fingerprints. The extracted features are unique and specific to each image/video. The visual hash is usually a global fingerprinting technique with crypto-system constraints. In this paper, we propose an innovative video content identification process which combines a visual hash function and a local fingerprinting. Thanks to a visual hash function, we observe the video content variation and we detect key frames. A local image fingerprint technique characterizes the detected key frames. The set of local fingerprints for the whole video summarizes the video or fragments of the video. The video fingerprinting algorithm identifies an unknown video or a fragment of video within a video fingerprint database. It compares the local fingerprints of the candidate video with all local fingerprints of a database even if strong distortions are applied to an original content.

Robust image hashing defines a feature vector that characterizes the image, independently of non-significant distortions of its content. As a consequence, the comparison between robust image hash vectors is able to indicate whether the corresponding images are equivalent or not, independently of visually non-significant distortions due for example to compression or re-sampling. We define a robust image hash based on radial projections of the image pixels. Specifically, our proposed radial hASH (RASH) considers moments of different orders to describe the luminance pdf of the pixels encountered on a set of lines articulated around the center of the image. In short, each RASH component is defined based on the moment of the pixels belonging to a specific line. Our paper provides a careful analysis of the robustness and discriminating capabilities of the RASH vectors computed based on different moment orders. As a first contribution, it demonstrates that the second order moment, i.e. the variance of the pixels on a line, allows for optimal trade-offs between the robustness and the discriminating capabilities of the resulting RASH vector. As a second contribution, extensive simulations prove that a decision engine based on RASH vectors cross-correlations is able to successfully identify pairs of equivalent or distinct images. Bottom line, the RASH assets are a low computational complexity, a strong robustness to both filtering and geometrical distortions, and a risk of collision that is estimated to less than 10 per million of images.

This paper presents a hardware implementation of a decoder for Digital Cinema images. This decoder enables us to deal with image size of 2K with 24 frames per second and 36 bits per pixels. It is the first implementation known nowadays that perfectly fits in one single Virtex-II® FPGA and includes AES decryption, JPEG 2000 decompression and fingerprinting blocks. This hardware offers therefore high-quality image processing as well as robust security.

Watermarking is largely used for copyright protection and fast search of images in databases. Another method for securely identifying images is to use hash functions. Digital signature standard, used in cryptosystem to dispute authentication documents, is based on hash functions. A digital signature is a bit stream dependent on key and content of document. For each document, the digital signature algorithm provides a unique output bit stream. In order to be efficient in images, the digital signature should be different if and only if the image content, and not the input bit stream, is different. Our new method is a one-way function for images. Using the radon transform and principal component analysis, we extract characteristics robust against geometrical transformation (rotation and scaling) and image processing attacks (compression, filtering, blurring).

We are present an algorithm for watermarking digital images with the aim of obtaining a very high quality of the watermarked image and a strong resistance to print and scan processes. We propose a method which combines an additive watermarking algorithm in the spatial domain providing resistance against cropping and exhaustive search and a synchronisation template in the Fourier domain providing resistance against geometrical deformations. The additive watermark in the spatial domain is based on an original generalized 2-D cyclic patterns for secret message embedding. The cyclic property and pattern redundancy facilitate detection and synchronisation against cropping and image processing basic attacks (like compression, filtering, blurring). This algorithm has to be complemented by a template insertion for providing resistance against rotation and scaling which are caused by the print and scan processes. We generate the template in the Fourier domain inserting some points locally. The watermark and the template are weighted by a human visual masking function. The global scheme, though very classical in its global concept, provides very efficient protection to a digital image which could be delivered both in a digital high quality format and in a printed form.

The Elodi project described aims at realizing a framework to produce a maintenance tool (diagnose and repair) of a complex system using the design data. This study is applied to a large scale and complex system: the high speed signalling system TVM 430. Since 1993, this device is used to bring the on-board signalling of the TGV (French high speed train). Currently, the only tools available are paper documents describing the system configuration, functionalities, and generic maintenance guide. The project goal is to combine these informations in one application. It gives diagnostic and repair support based on a model and data observations. The TVM system is presented following by a description of the problem of the model-based diagnoses. Finally, the modelling problem to obtain a tool for the maintenance task is discussed.