Video Signal-Dependent Noise Estimation via Inter-Frame Prediction

Li, Yanhao; Gardella, Marina; Bammey, Quentin; Nikoukhah, Tina; Gioi, Rafael Grompone von; Colom, Miguel; Morel, Jean‐Michel

doi:10.1109/icip46576.2022.9897605

Cited by 4 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a consequence, current synthetic image detection methods, including the proposed one, should still be considered a research artefact, and not be used as proof that an image is actually forged. For practical usability, setting an automatic threshold would be crucial, for instance with a contrario analysis [18], [19], a promising approach in forensics [2]- [6], [26], [36], [42], [44] Finally, we note that the proposed method is trained on diffusion-model-generated, photorealistic images. It is not trained to work on GAN images, for which numerous tools already exists.…”

Section: Discussion and Limitationsmentioning

confidence: 99%

Synthbuster: Towards Detection of Diffusion Model Generated Images

Bammey

2023

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

Synthetically-generated images are getting increasingly popular. Diffusion models have advanced to the stage where even non-experts can generate photo-realistic images from a simple text prompt. They expand creative horizons but also open a Pandora's box of potential disinformation risks. In this context, the present corpus of synthetic image detection techniques, primarily focusing on older generative models like Generative Adversarial Networks, finds itself ill-equipped to deal with this emerging trend. Recognizing this challenge, we introduce a method specifically designed to detect synthetic images produced by diffusion models. Our approach capitalizes on the inherent frequency artefacts left behind during the diffusion process. Spectral analysis is used to highlight the artefacts in the Fourier transform of a residual image, which are used to distinguish real from fake images. The proposed method can detect diffusion-model-generated images even under mild JPEG compression, and generalizes relatively well to unknown models. By pioneering this novel approach, we aim to fortify forensic methodologies and ignite further research into the detection of AI-generated images.

show abstract

Section: Discussion and Limitationsmentioning

confidence: 99%

Synthbuster: Towards Detection of Diffusion Model Generated Images

Bammey

2023

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

show abstract

“…The overall pipeline is shown in Figure 1. A preliminary version of this method was described in [14].…”

Section: Proposed Methodsmentioning

confidence: 99%

A Signal-dependent Video Noise Estimator Via Inter-frame Signal Suppression

Li,

Gardella,

Bammey

et al. 2023

Image Processing on Line

View full text Add to dashboard Cite

We propose a block-based signal-dependent noise estimation method on videos, that leverages inter-frame redundancy to separate noise from signal. Block matching is applied to find block pairs between two consecutive frames with similar signal. Then the Ponomarenko et al. method is extended to video by sorting pairs by their low-frequency energy and estimating noise in the high frequencies. Experiments on a real dataset of drone videos show its performance for different parameter settings and different noise levels. Two extensions of the proposed method using subpixel matching and for multiscale noise estimation are respectively analyzed. Source CodeThe reviewed source code and documentation for this algorithm are available from the web page of this article 1 . Compilation and usage instructions are included in the README.md file of the archive.

show abstract

“…Based on the non-accidentalness principle, this theory proposes to detect data based on their unlikelihood under a background hypothesis, by thresholding the results based on a tolerated limit on the number of false alarms (NFA) under the hypothesis. This paradigm has seen successful applications in varied detection tasks [1,21,22,23,24,28,29,30], including forensics [3,8,10,5,10,17,19,32,29].…”

Section: Introductionmentioning

confidence: 99%

A Contrario Mosaic Analysis for Image Forensics

Bammey

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

With the advent of recent technologies, image editing has become accessible even without expertise. However, this ease of manipulation has given rise to malicious manipulation of images, resulting in the creation and dissemination of visually-realistic fake images to spread disinformation online, wrongfully incriminate someone, or commit fraud. The detection of such forgeries is paramount in exposing those deceitful acts. One promising approach involves unveiling the underlying mosaic of an image, which indicates in which colour each pixel was originally sampled prior to demosaicing. As image manipulation will alter the mosaic as well, exposing the mosaic enables the detection and localization of forgeries. The recent introduction of positional learning has facilitated the identification of the image mosaic. Nevertheless, the clues leading to the mosaic are subtle and frail against common operation such as JPEG compression. The pixelwise estimation of the mosaic is thus often imprecise, and a comprehensive analysis and aggregation of the results are necessary to effectively detect and localize forged areas. In this work, we propose mimic: Mosaic Integrity Monitoring for Image a Contrario forensics. an a contrario method to analyse a pixelwise mosaic estimation. We show that despite the weakness of these traces, the sole analysis of mosaic consistency is enough to beat the state of the art in forgery detection and localization on uncompressed images. Moreover, results are promising even on slightly-compressed images. The a contrario framework ensures robustness against false positives, and the complementary nature of mosaic consistency analysis to other forensic tools makes our method highly relevant for detecting forgeries in high-quality images.

show abstract

Video Signal-Dependent Noise Estimation via Inter-Frame Prediction

Cited by 4 publications

References 25 publications

Synthbuster: Towards Detection of Diffusion Model Generated Images

Synthbuster: Towards Detection of Diffusion Model Generated Images

A Signal-dependent Video Noise Estimator Via Inter-frame Signal Suppression

A Contrario Mosaic Analysis for Image Forensics

Contact Info

Product

Resources

About