ALCORLab

ALCORLab computer vision research laboratory of the Department of Computer, Control, and Management Engineering A. Ruberti, Sapienza University of Rome

[Taiba Majid Wani @taiba_wani]

📢 Excited to share that our paper "ABC-CapsNet: Attention-Based Cascaded Capsule Network for Audio Deepfake Detection" has been accepted at @WiCVworkshop @CVPR in Seattle, USA, ranking in the top 25% of submissions! 🎉

[Luca Maiano @lucamaiano]

Last Sunday, I had the pleasure of presenting our latest work at #ROMCIR2024, the #ECIR2024 Credible Information Retrieval workshop. I introduced a multimodal deepfake detector that analyzes the audio and video parts to recognize fake conten.

@alcor_lab

[Irene Amerini @ireclimb]

Huge congratulations to the new PhD graduate @lucamaiano for successfully defending his PhD thesis about multimedia forensics and multimodal fake news detection today. We have a new Doctor in Alcor Lab! @alcor_lab @DIAGSapienza

[WiCV @WiCVworkshop]

Call for Women in Computer Vision #WiCV workshop organizer at #ECCV2024
If you want to be an organizer for the #WiCV workshop at ECCV 2024, please fill out the form by Feb 04, 2024:
https://t.co/RQhnMcjiMM
Please RT and share. We look forward to your application.

[ALCORLab @alcor_lab]

The special session "Artificial Intelligence for Countering Disinformation and Information Warfare" organized in conjuction with #WCCI2024 #IJCNN is now accepting submission. 15 January 2024 Paper Submission Deadline. https://t.co/7x62r0EcM7

[ALCORLab @alcor_lab]

📢 Yesterday during #ACMMM2023 - 5th Workshop on Analysis, Understanding and Promotion of Heritage Contents Matteo Zaramella presented our work "Why Don’t You Speak?: A Smartphone Application to Engage Museum Visitors Through Deepfakes Creation" and won the best paper award! 🏆

[ALCORLab @alcor_lab]

Next week at BiDS will be presented our joint work with European Space Agency - ESA #philab on Diffusion Models for Earth Observation use cases. Thanks to Bertrand Le Saux, Fulvio Sanguigni, Mikolaj Czerkawski, Lorenzo Papa and Irene Amerini #alcorlab #AI4EO

[ALCORLab @alcor_lab]

https://t.co/tPhRjXgf1O

[ALCORLab @alcor_lab]

Today great brainstorming at #DIAG with #alcorlab researchers on many topic from path tracing to AI-generated images detection and ground-to-aerial image matching!https://t.co/VPMr7IzbGR

[ALCORLab @alcor_lab]

https://t.co/lqR3v5p5fD

[ALCORLab @alcor_lab]

https://t.co/biWqyFzMwj

[Lamberto Ballan @lambertoballan]

Classmates... a few years later :) #ICIAP23 @giuseppeserra @ireclimb

[ALCORLab @alcor_lab]

@florian_merkle @ireclimb

[Irene Amerini @ireclimb]

#ICIAP2023 Udine with the keynote by Andrew Fitzgibbon. We will present two papers as @alcor_lab on adversarial machine learning and deepfake audio detection this afternoon.

[Luca Maiano @lucamaiano]

Glad I was at #ICVSS2023 this week. It was a week full of very interesting meetings and lessons with the best #computervision experts

[Irene Amerini @ireclimb]

At Warwick University to discuss about deepfake detection to counteract misinformation. Nice discussions and good weather. Thanks Victor Sanchez for the invitation and to all to the SIPLab @WarwickDCS @alcor_lab

[Luca Maiano @lucamaiano]

On Monday, we presented our latest work at the 2023 24th International Conference on
Digital Signal Processing (DSP) in Greece.

The paper presents a method for ground-to-aerial matching for forensic applications.

[ALCORLab @alcor_lab]

The IAPR Computational Forensics TC will host a virtual seminar on 11 May at 9 am in Rome, 10 am in London, UK. The seminar will be given by Dr Benedetta Tondi from University of Siena on "Backdooring Deep Neural Networks: from threats to opportunities.” https://t.co/BvNBkngVKN

[Luca Maiano @lucamaiano]

Thanks to my colleagues @lorenzopapa597, Lorenzo Faiella, Luca Corvitto, and @ireclimb for their collaboration

Special thanks to Lorenzo Papa for presenting our paper today.

[Luca Maiano @lucamaiano]

Glad to share our latest work at #IWBF2023.

#stablediffusion can generate highly realistic images that can be used to spread #disinformation. We propose a comprehensive analysis of this phenomenon by studying how difficult it is to create and recognize realistic fake images.