The Face Deepfake Detection Challenge.

Multimedia data manipulation and forgery has never been easier than today, thanks to the power of Artificial Intelligence (AI). AI-generated fake content, commonly called Deepfakes, have been raising new issues and concerns, but also new challenges for the research community. The Deepfake detection task has become widely addressed, but unfortunately, approaches in the literature suffer from generalization issues. In this paper, the Face Deepfake Detection and Reconstruction Challenge is described. Two different tasks were proposed to the participants: (i) creating a Deepfake detector capable of working in an "in the wild" scenario; (ii) creating a method capable of reconstructing original images from Deepfakes. Real images from CelebA and FFHQ and Deepfake images created by StarGAN, StarGAN-v2, StyleGAN, StyleGAN2, AttGAN and GDWCT were collected for the competition. The winning teams were chosen with respect to the highest classification accuracy value (Task I) and "minimum average distance to Manhattan" (Task II). Deep Learning algorithms, particularly those based on the EfficientNet architecture, achieved the best results in Task I. No winners were proclaimed for Task II. A detailed discussion of teams' proposed methods with corresponding ranking is presented in this paper.

Hybrid reference-based Video Source Identification.

Millions of users share images and videos generated by mobile devices with different profiles on social media platforms. When publishing illegal content, they prefer to use anonymous profiles. Multimedia Forensics allows us to determine whether videos or images have been captured with the same device, and thus, possibly, by the same person. Currently, the most promising technology to achieve this task exploits unique traces left by the camera sensor into the visual content. However, image and video source identification are still treated separately from one another. This approach is limited and anachronistic, if we consider that most of the visual media are today acquired using smartphones that capture both images and videos. In this paper we overcome this limitation by exploring a new approach that synergistically exploits images and videos to study the device from which they both come. Indeed, we prove it is possible to identify the source of a digital video by exploiting a reference sensor pattern noise generated from still images taken by the same device. The proposed method provides performance comparable with or even better than the state-of-the-art, where a reference pattern is estimated from video frames. Finally, we show that this strategy is effective even in the case of in-camera digitally stabilized videos, where a non-stabilized reference is not available, thus solving the limitations of the current state-of-the-art. We also show how this approach allows us to link social media profiles containing images and videos captured by the same sensor.

Mixed-Valence Porphyrin π-Cation Radical Derivatives: Electrochemical Investigations.

The electrochemistry of [Cu(OEP)] and [Ni(OEP)] are compared with the mixed-valence π-cations [Cu(OEP•/2)]2+and[Ni(OEP•/2)]2+. These electrochemical studies, carried out with cyclic voltametry and hydrodynamic voltametry, show that the mixed valence π-cations have distinct electrochemical properties, although the differences between the [M(OEP)](+/0) and [M(OEP•/2)]2+/0 processes are subtle.

Ferrocenyl-functionalised terpyridines and their transition-metal complexes: syntheses, structures and spectroscopic and electrochemical properties.

Terpyridine ligands of the type Fc'-X-tpy (Fc'=ferrocenyl or octamethylferrocenyl, X=rigid spacer, tpy'=4'-substituted 2,2':6',2''-terpyridine) were prepared, crystallographically characterised and used for the synthesis of di- and trinuclear bis(terpyridine) complexes of RuII, FeII and ZnII. Donor-sensitiser dyads and triads based on RuII were thoroughly investigated by (spectro)electrochemistry, UV/Vis, transient absorption and luminescence spectroscopy, and an energy level scheme was derived on the basis of the data collected. Intramolecular quenching of the photoexcited RuII complexes by the redox-active Fc' groups can occur reductively and by energy transfer. Both the redox potential of the donor Fc' and the nature of the spacer X have a decisive influence on excited-state lifetimes and emission properties of the complexes. Some of the compounds show room-temperature luminescence, which is unprecedented for ferrocenyl-functionalised compounds of this kind.

Molecular structure, solution chemistry and biological properties of the novel [ImH][trans-IrCl(4)(Im)(DMSO)], (I) and of the orange form of [(DMSO)(2)H][trans-IrCl(4)(DMSO)(2)], (II), complexes.

The new iridium(III) complex, imidazolium[trans(DMSO,imidazole)tetrachloroiridate(III)], (I) (DMSO=dimethyl sulfoxide), and the orange form of [(DMSO)(2)H][trans(DMSO)(2)tetrachloroiridate(III)], (II) have been prepared and characterized, both in the solid state and in solution, by X-ray diffraction and by various physicochemical techniques. Single crystal X-ray diffraction studies point out that complex (II) is isomorphous to the ruthenium(III) analogue, [(DMSO)(2)H][trans-RuCl(4)(DMSO)(2)], (III). Crystallographic data are the following: a=16.028(2) A, b=24.699(3) A, c=8.262(1) A, in space group Pbca (Z=8) for (imidazolium)[trans(DMSO,imidazole)tetrachloroiridate(III)], (I); and a=9.189(2) A, b=16.511(4) A, c=14.028(3) A, beta=100.82(2) degrees in space group P2/n (Z=4) for [(DMSO)(2)H][trans(DMSO)(2)tetrachloroiridate(III)], (II). Visible absorption spectra show that both complexes are stable for several days, at pH 7.4, at room temperature. No significant chloride hydrolysis is observed, even at high temperature (70 degrees C), over 24 h. The extreme stability of these iridium(III) complexes within a physiological buffer was further assessed by (1)H NMR; in addition, cyclic voltammetry measurements evidenced a high stability of the oxidation state +3. Preliminary biological studies show that both complexes do not bind appreciably bovine serum albumin nor inhibit significantly the proliferation of representative human tumor cell lines, suggesting that hydrolysis of coordinated chlorides is a crucial feature for the biological properties and the antitumor activity of the parent ruthenium(III) complexes.

Design, synthesis, and physicochemical and biological characterization of a new iron chelator of the family of hydroxychromenes.

Increasing evidence suggests that iron plays an important role in tissue damage both during chronic iron overload diseases (i.e., hemochromatosis) and when, in the absence of actual tissue iron overload, iron is delocalized from specific carriers or intracellular sites (inflammation, neurodegenerative diseases, postischaemic reperfusion, xenobiotic intoxications, etc.). In the present work, we appropriately modified an iron chelator of the hydroxychromene family in order to obtain a tridentate chelator that would inactivate the iron redox cycle after its complexation, with a view to using this molecule in human therapy and/or in disease prevention. We synthesized such a chelator for the first time and show, by different physicochemical analysis, its tridentate nature and, importantly, its capacity to chelate iron with enough strength to inhibit both iron-dependent H(2)O(2) generation and lipid peroxidation in in vitro biological systems.