First Vector Graphic

Portions of the training set are now available!


Training set are now available!


Test set are now available!








Main Contact : Luca Guarnera luca.guarnera@unict.it

App
Challenge Description

Since their first creation, Deepfakes have been widely used for malicious purposes such as in the pornography industry. Therefore, the need to counteract the illicit use of this powerful technology was born. So much work has been proposed in the literature but unfortunately most of it lacks generalizability. The purposes of this challenge is to create (a) "in the wild" Deepfake detection algorithms that can counteract the malicious use of these powerful technologies and (b) try to reconstruct the source image from those Deepfakes. To this aim, the challenge will be divided into two tasks as described in the following subsections. Given the importance and the dangerousness of Deepfake the entire challenge will focus only on Deepfake images of human faces.

TASK 1: DEEPFAKE DETECTION TASK

This part of the challenge is the classical Deepfake detection binary classification task. The participants’ solutions should be able to define whether an image is real or deepfake. The participants’ solutions will be evaluated with particular emphasis in terms of “robustness” to common alterations on images such as: rotation, mirroring, gaussian-filtering, scaling, cropping and re-compressions.
The winning team of Task 1 will receive a prize of 500 €.

TASK 2: SOURCE IMAGE RECONSTRUCTION

This part of the challenge is a task never addressed in literature: given a deepfake image generated with a specific architecture and model (StarGAN-v2 [8]), the goal is to best reconstruct the source image in its original form starting from the deepfake counterpart.
The winning team of Task 2 will receive a prize of 500 €.

Additional details of the tasks, datasets, and the challenge rules will be described in Evaluation Criteria, Competitions Rules and Dataset Sections.



Important Dates
FInal results on the test-set will be disclosed at conference time.

Data Provided and Evaluation Criteria

The proposed challenge aims to ask the participants to produce new techniques to fight against deepfake images.

For this purpose the challenge will be divided into two tasks with different objectives and submissions evaluation metrics.

The two tasks will be described in terms of data provided and evaluation metrics as the following.

App
TASK 1: DEEPFAKE DETECTION TASK

The dataset provided will be composed of both real and Deepfake images of human faces. The Deepfake images will be generated by means of several GAN architectures based on well-known deepfake manipulations such as: face transfer, face swap and style transfer. The training set will be composed of images and organized into several ZIP files having as structure "LABEL-GANname.ZIP" (e.g., "0-CELEBA.ZIP", "1-StarGAN.ZIP"), where LABEL is the Ground Truth (value equal to 0 if the dataset contains real images; value equal to 1 if the dataset contains deepfake images).
Participants will organize these datasets as they see fit (split them into training and validation, define the split percentage, etc.) and can perform any augmentation operation as long as they use only the dataset provided for the competition.
The test set, released in the last part of the competition (see the Important Dates section), will be a TEST.ZIP file composed by several real and Deepfake images similar to those of the training set, and in addition, images obtained by applying some processing (rotation, mirroring, gaussian-filtering, scaling, cropping and re-compression) will be introduced.
For this task, the winning team will be selected with respect to the highest classification accuracy value obtained on the Real vs. Deepfake binary classification task computed with participants’ solutions on a further set of data not available until the end of the challenge submission period.
NOTE THAT: only the solutions that will be submitted not later than the defined deadline will be considered. All solutions submitted after the deadline will not be considered.

Figure 1 summarizes the objective of Task 1.

App

Figure 1: Deepfake Detection task.

TASK 2: SOURCE IMAGE RECONSTRUCTION

The dataset provided will consist of images manipulated by the Stargan-v2 [8] architecture. In detail, each Deepfake sample is obtained through the attribute manipulation operation performed via the StarGAN v2 architecture on a source image (src) with respect to the attributes of a reference image (ref). Figure 2 shows an example.
The dataset is organized into 3 different ZIP files: SOURCES.ZIP, REFERENCES.ZIP and Deepfake.ZIP. Each Deepfake sample, available in Deepfake.ZIP, has a filename that follows the following structure: deepfake-src_IDs-ref_IDr.JPG, where IDs refers to the ID of the source image that can be found in the SOURCES.ZIP (with filename src_IDs.JPG) and, IDr refers to the ID of the reference image that can be found in the REFERENCES.ZIP with filename ref_IDr.JPG). See Figure 2.
Participants will organize these datasets as they see fit (split them into training and validation, define the split percentage, etc.) and can perform any augmentation operation as long as they only use the dataset provided for the competition.
The test set, released in the last part of the competition (see the Important Dates section), will consist ONLY of Deepfake images. Participants must share only the reconstructed images.
For this competition, the winning team will be selected based on the "minimum average distance to Manhattan" calculated (by the organizers) between the sources (available only to the organizers and made public once the competition is over) and the images reconstructed by the participants.
NOTE THAT: only the solutions that will be submitted not later than the defined deadline will be considered. All solutions submitted after the deadline will not be considered.

Figure 3 summarizes the objective of Task 2.

App

Figure 2: Name structure of source images, reference images and deepfake images.



App

Figure 3: Source image reconstruction task.



Competitions Rules
(registration, submission, presentation)

Participants should start preparing a descriptive documentation for their final submission. Indeed, top-ranked teams will be invited to present their works at the ICIAP 2021 conference during the competition session. More details will be available in this web site.
Each team is invited to participate freely in one of the two tasks or in all the tasks listed above.
No later than 15/01/2022 participants must register at this following website where they must specify the team name, affiliation, and other informations.
It will not be allowed to unify teams after the beginning of the challenge even if they belong to the same university.
The technical documentation that participants must submit no later than 28/02/2022 must be written in English and must contain all the details of the proposed approach (for example, if you use a deep neural network algorithm then you must describe the architecture used, the parameters, etc.) and the results obtained, as well as the comparison with the methods reported at the Related Research section and other recent state-of-the-art methods that are deemed necessary, inherent, and important to participants.

TASK 1: SPECIFIC RULES

For the first task, by using the TEST set released on the date indicated in Important Dates section, participants will need to create and submit a simple .TXT file specifying for each row the name of the analyzed image and the estimated label:
--- file.txt
------ name_image_1.jpg     0/1 \n
------ name_image_2.jpg     0/1 \n
...........................................................
------ name_image_N.jpg     0/1

Example:
--- team0.txt
------ 000.jpg 1 \n
------ 001.jpg 1 \n
------ 002.jpg 0 \n
...........................................................
------ N.jpg 0

The winner will be determined based on the highest classification accuracy value obtained in this last phase.

NOTE THAT participants must upload on THIS WEBSITE (after having carried out the login procedure) and ONLY for TASK 1 the TXT file described above and the related documentation (PDF) written in English language. The data upload phase on this platform will be made available on the dates reported in the Important Dates section. Participants have only 5 different uploads available: one is used to upload the documentation (.pdf) and four are available to upload the .TXT of the results. Only the last files (.TXT and documentation) uploaded will be considered by the organizers. Results without documentation (.pdf) will not be evaluated by the organizers.
Training is allowed only on the provided datasets. Participants are allowed to fine-tune pre-trained models among those available at:

TASK 2: SPECIFIC RULES

For the second task, by using the TEST set released on the date indicated in Important Dates section, participants must share a Google Drive folder, with the related documentation (PDF) and a sub-folder containing ONLY the reconstructed images (with the same name as the respective Deepfake image), to the following email address: deepfakechallenge@gmail.com. The email must have as subject TEAM_NAME - ICIAP2022_Task2

The winner will be determined based on the lowest mean value of Manhattan distances.
Results without documentation (.pdf) will not be evaluated by the organizers.

Training is allowed only on the provided datasets. Participants are allowed to fine-tune pre-trained model among those available at:


Dataset
TASK 1: DEEPFAKE DETECTION TASK

Two datasets of real face images were used for the employed experimental phase: CelebA and FFHQ. Different Deepfake images were generated considering StarGAN, GDWCT, AttGAN, StyleGAN and StyleGAN2 architectures. In particular, CelebA images were manipulated using pre-trained models available on Github, taking into account StarGAN, GDWCT and AttGAN. Images of StyleGAN and StyleGAN2 created through FFHQ were downloaded as detailed in the following:

  • [a] CelebA [1] (CelebFaces Attributes Dataset): a large-scale face attributes dataset with more than 200 K celebrity images, containing 40 labels related to facial attributes such as hair color, gender and age. The images in this dataset cover large pose variations and background clutter. The dataset is composed by 178 × 218 JPEG images.
  • [b] FFHQ [2] (Flickr-Faces-HQ): is a high-quality image dataset of human faces with variations in terms of age, ethnicity and image background. The images were crawled from Flickr and automatically aligned and cropped using dlib [43]. The dataset is composed by high-quality 1024 × 1024 PNG images.
  • [c] StarGAN [3] is able to perform Image-to-image translations on multiple domains using a single model. Using CelebA as real images dataset, every image was manipulated by means of a pre-trained model obtaining a final resolution equal to 256 × 256.
  • [d] GDWCT [4] is able to improve the styling capability. Using CelebA as real images dataset, every image was manipulated by means of a pre-trained model obtaining a final resolution equal to 216 × 216.
  • [e] AttGAN [5] is able to transfers facial attributes with constraints. Using CelebA as real images dataset, every image was manipulated by means of a pre-trained model obtaining a final resolution equal to 256 × 256.
  • [f] StyleGAN [6] is able to transfers semantic content from a source domain to a target domain characterized by a different style. Images have been generated considering FFHQ as dataset in input with 1024 × 1024 resolution .
  • [g] StyleGAN2 [7] improves STYLEGAN quality with the same task. Images have been generated considering FFHQ as dataset in input with 1024 × 1024 resolution .

TASK 2: SOURCE IMAGE RECONSTRUCTION

  • [a] CelebA [1] (CelebFaces Attributes Dataset): a large-scale face attributes dataset with more than 200 K celebrity images, containing 40 labels related to facial attributes such as hair color, gender and age. The images in this dataset cover large pose variations and background clutter. The dataset is composed by 178 × 218 JPEG images.
  • [b] StarGAN-v2 [8] is able to perform Image-to-image translations on multiple domains using a single model. Using CelebA as real images dataset, every image was manipulated by means of a pre-trained model obtaining a final resolution equal to 256 × 256.

REFERENCES

  1. [1] Z. Liu, P. Luo, X. Wang and X. Tang, Deep Learning Face Attributes in the Wild, 2015 IEEE International Conference on Computer Vision (ICCV), 2015, pp. 3730-3738, doi: 10.1109/ICCV.2015.425.
  2. [2] https://github.com/NVlabs/ffhq-dataset, accessed on 03/11/2021.
  3. [3] Y. Choi, M. Choi, M. Kim, J. Ha, S. Kim and J. Choo, StarGAN: Unified Generative Adversarial Networks for Multi-domain Image-to-Image Translation, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 8789-8797, doi: 10.1109/CVPR.2018.00916.
  4. [4] W. Cho, S. Choi, D. K. Park, I. Shin and J. Choo, Image-To-Image Translation via Group-Wise Deep Whitening-And-Coloring Transformation, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 10631-10639, doi: 10.1109/CVPR.2019.01089.
  5. [5] Z. He, W. Zuo, M. Kan, S. Shan and X. Chen, AttGAN: Facial Attribute Editing by Only Changing What You Want, in IEEE Transactions on Image Processing, vol. 28, no. 11, pp. 5464-5478, Nov. 2019, doi: 10.1109/TIP.2019.2916751.
  6. [6] T. Karras, S. Laine and T. Aila, A Style-Based Generator Architecture for Generative Adversarial Networks, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 4396-4405, doi: 10.1109/CVPR.2019.00453.
  7. [7] T. Karras, S. Laine, M. Aittala, J. Hellsten, J. Lehtinen and T. Aila, Analyzing and Improving the Image Quality of StyleGAN, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 8107-8116, doi: 10.1109/CVPR42600.2020.00813.
  8. [8] Y. Choi, Y. Uh, J. Yoo and J. -W. Ha, StarGAN v2: Diverse Image Synthesis for Multiple Domains, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 8185-8194, doi: 10.1109/CVPR42600.2020.00821.


Related Research
https://iplab.dmi.unict.it/mfs/Deepfakes/


  • - Giudice, O.; Guarnera, L.; Battiato, S. Fighting Deepfakes by Detecting GAN DCT Anomalies. J. Imaging 2021, 7, 128. https://doi.org/10.3390/jimaging7080128. PDF, WEB SITE
  • - L. Guarnera, O. Giudice and S. Battiato, Fighting Deepfake by Exposing the Convolutional Traces on Images, in IEEE Access, vol. 8, pp. 165085-165098, 2020, doi: 10.1109/ACCESS.2020.3023037. PDF, WEB SITE
  • - L. Guarnera, O. Giudice and S. Battiato, DeepFake Detection by Analyzing Convolutional Traces, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2020, pp. 2841-2850, doi: 10.1109/CVPRW50498.2020.00341. PDF, WEB SITE
  • - L. Guarnera, O. Giudice, C. Nastasi and S. Battiato, Preliminary Forensics Analysis of DeepFake Images, 2020 AEIT International Annual Conference (AEIT), 2020, pp. 1-6, doi: 10.23919/AEIT50178.2020.9241108. PDF, WEB SITE


Awards

TASK 1: DEEPFAKE DETECTION TASK

The winning team of Task 1 will receive a prize of 500 €.



TASK 2: SOURCE IMAGE RECONSTRUCTION

The winning team of Task 2 will receive a prize of 500 €.

App


Publication plan
The best selected solutions will be described in a challenge report paper. The paper will include the most significant works and their findings. In addition to the ICIAP 2021 challenge section, the authors of the best selected works will be invited to submit their contribution to a special issue of a valuable Journal.


Organizers
Sebastiano Battiato, Full Professor, University of Catania, Italy
Email: battiato@dmi.unict.it
Google Scholar: https://scholar.google.it/citations?hl=it&user=OplbtHgAAAAJ
Webpage: http://www.dmi.unict.it/~battiato/


Oliver Giudice (Ph.D.), Researcher, Banca d’Italia, Rome, Italy
IEEE Member
Email: oliver.giudice@bancaditalia.it
Google Scholar: https://scholar.google.com/citations?user=YrVdU3IAAAAJ&hl=it&oi=ao

Francesco Guarnera (Ph.D. Student), University of Catania, Italy
Email: francesco.guarnera@unict.it
Google Scholar: https://scholar.google.com/citations?user=DyGUX9IAAAAJ&hl=it&oi=ao

Luca Guarnera (Ph.D.), Fellow Researcher, University of Catania, Italy
IEEE Member (IEEE Signal Processing Society Member)
Email: luca.guarnera@unict.it
Google Scholar: https://scholar.google.com/citations?user=x1Zhq3gAAAAJ&hl=it
Webpage: https://www.dmi.unict.it/lguarnera/

Alessandro Ortis, Postdoctoral Researcher, University of Catania, Italy
Email: ortis@dmi.unict.it
Google Scholar: https://scholar.google.it/citations?user=gcztqXgAAAAJ
Webpage: http://www.dmi.unict.it/~ortis/

Antonino Paratore, Digital Forensics Expert, iCTLab srl.
Email: antonino.paratore@ictlab.srl
Google Scholar: https://scholar.google.com/citations?user=Ilbm2RYAAAAJ&hl=it&oi=ao

Giovanni Puglisi, Associate Professor, University of Cagliari, Italy
Email: puglisi@unica.it
Webpage: https://unica.it/unica/it/ateneo_s07_ss01.page?contentId=SHD30978

Sponsor
...