GAP: Differentially Private Graph Neural Networks with Aggregation Perturbation
A tailored Graph Neural Network architecture with Differential Privacy guarantees for both training and inference.
A Centre of Excellence delivering next generation AI Research and Training at the service of Media, Society and Democracy.
Motivated by the challenges, risks and opportunities that the wide use of AI brings to media, society and politics, AI4Media aspires to become a centre of excellence and a wide network of researchers across Europe and beyond, with a focus on delivering the next generation of core AI advances to serve the key sector of Media, to make sure that the European values of ethical and trustworthy AI are embedded in future AI deployments, and to reimagine AI as a crucial beneficial enabling technology in the service of Society and Media.
The AI4Media consortium, comprising 30 leading partners in the areas of AI and media (9 universities, 9 research centres, and 12 industrial partners) and 35 associate members, will establish the networking infrastructure to bring together the currently fragmented European AI landscape in the field of media, and foster deeper and long-running interactions between academia and industry, including Digital Innovation Hubs. It will also shape a research agenda for media AI research, and implement research and innovation both with respect to cutting-edge technologies at the core of AI research, and within specific fields of media-related AI. AI4Media will provide a targeted funding framework through open calls, to speed up the uptake of innovations developed within the network. A PhD programme will further enhance links to the industry and the fostering and exchange of talent, while providing motivation to prevent brain drain, and a set of use cases will be developed by the network to demonstrate the impact of the achieved advances in the media sector. The Excellence Centre that is established during the AI4Media project, and the ecosystem that will grow around it, will provide a long-term basis for the support of AI excellence in Europe, long after the project ends, with the aim of ensuring that Ethical AI guided by European values assumes a global leading role in the field of Media.
A tailored Graph Neural Network architecture with Differential Privacy guarantees for both training and inference.
Python implementation of novel Cycle In Cycle Generative Adversarial Network (C2GAN) for the task of keypoint-guided image generation. The C2GAN is a cross-modal framework exploring a joint exploitation of the keypoint and the image data in an interactive...
A tool to allow Visual Transformers (VTs) to learn spatial relations within an image making the VT training much more robust when training data is scarce. The tool can be used jointly with the standard (supervised) training and it does not depend on speci...
PyTorch implementation of Intrinsic-Extrinsic Preserved Generative Adversarial Network (IEP-GAN) for both intrinsic (i.e., shape) and extrinsic (i.e., pose) information preservation. Extrinsically, a co-occurrence discriminator is used to capture the stru...
A holistic learning framework for Novel Class Discovery (NCD), which adopts contrastive learning to learn discriminate features with both the labeled and unlabeled data. The Neighborhood Contrastive Learning (NCL) framework effectively leverages the local...
PyTorch implementation of a Geometry-Contrastive Transformer for Generalized 3D Pose Transfer. The novel GC-Transformer can freely conduct robust pose transfer on LARGE meshes at no cost, which could be a boost to Transformers in 3D fields.
Mini-batch trimming - a curriculum learning method for improving model generalization via progressively increasing the training difficulty
This is a fork of Few-shot Object Detection (FsDet) method, adding an easy to use tool for training on custom datasets
AdaFamily: A family of Adam-like adaptive gradient methods for training neural networks
Is “AI Science and Engineering” an un upcoming scientific discipline that can fuse AI, brain and mind studies and social engineering? As Artificial Intelligence (AI) studies and research flourish worldwide, it is worth debating whether we already observe ...
Our world is increasingly complex, in terms of both its material components (e.g., smart cities, infrastructure) and its social processes (e.g., social media outreach). Both individual humans and entire societies find it difficult to cope with world compl...
A novel unsupervised domain adaptation approach for action recognition from videos, inspired by recent literature on contrastive learning. It comprises a novel two-headed deep architecture that simultaneously adopts cross-entropy and contrastive losses fr...
Novel two-stage framework with a new Cascaded Cross MLP-Mixer (CrossMLP) sub-network in the first stage and one refined pixel-level loss in the second stage. In the first stage, the CrossMLP sub-network learns the latent transformation cues between image ...
A library of self-supervised methods for unsupervised visual representation learning powered by PyTorch Lightning. It aims at providing SotA self-supervised methods in a comparable environment while, at the same time, implementing training tricks. While t...
As the backward algorithm of SVD is prone to have numerical instability, we implement a variety of end-to-end SVD methods by manipulating the backward algortihms in this repository. They include: - SVD-Pad'e: use Pad'e approximants to closely approximate...
Novel framework for Playable Video Generation that is trained in a self-supervised manner on a large dataset of unlabelled videos. We employ an encoder-decoder architecture where the predicted action labels act as bottleneck. The network is constrained to...
Image acquisition. Mathematical modeling of image formation. Introduction to image processing and analysis. Camera calibration. Stereo vision. Depth estimation. Object localization. 3D image analysis. Surface geometry. Object topology. Object landmarks an...
Our dream to make machines sense and perceive (notably see) comes true: nowadays Computer Vision enables diverse applications: - Autonomous Systems (cars, drones, vessels) Perception, - Robotics Perception and Control, - Intelligent Human-Machine Int...
Diffprivlib is a general-purpose library for experimenting with, investigating and developing applications in, differential privacy.
Detect perceptually identical audio segments/matches among audio items or streams.
Nowadays, digital images and video are everywhere. Image Processing revolutionizes very many domains, notably: -Digital Media (video/image/movie) Content Production and Broadcasting, Social Media Analytics, -Medical/Biological/Dental Imaging and Diagn...
Many CVML scientists, engineers and enthusiasts do not have solid mathematical background, as it is so easy to jump into almost any CVML domain using available libraries and frameworks. This is very much true in Deep Learning and leads to a cacophony of i...
Nowadays, Artificial Intelligence drives scientific and economic growth worldwide. This is largely due to advances in Machine Learning (ML), notably in Deep Neural Networks (DNNs), which are essentially massive ‘learning by experience/examples’ systems. T...
Digital signals and images have great applications in Medical diagnosis and treatment, as well as in Biology, Biomedical Engineering, Dentistry and Pharmacology studies.
Acoustic digital signals are ubiquitous and have many applications in several disciplines, notably in: -Digital Media, Social Media (music, voice signals), -Biomedical Signal Analysis and Diagnosis, -Scientific signal acquisition of any sort, e.g., E...
Human Centered Computing has very many application areas in, e.g.,: -Human-Computer Interfaces and Human-Robot Interfaces -Social Media Analytics -Video Analysis -Affective Computing -Biometrics -Privacy Protection
Nowadays, digital videos are everywhere and revolutionize very many domains, notably: -Digital Media (video/movie) Content Production and Broadcasting, -Social Media Streaming and Analytics (g., YouTube), -Mobile computing and streaming -Videoconfer...
Automatic Control and Robotics have great applications in Industry, notably in assembly lines and the automation of industrial processes, as well as in medicine and in assisted living. In particular, companion robots are very timely and get a lot of atten...
Autonomous Marine Vessels come in many forms that have many applications, e.g.,: -Autonomous Ships -Autonomous Sailing Boats -Autonomous Underwater Vessels
Autonomous Systems (AS) have nowadays a multitude of applications: -Autonomous Cars -Autonomous Drones -Autonomous Marine Systems (surface, underwater) -Autonomous Robots. In particular, Autonomous Cars and Autonomous Drones and Drone Swarms are ve...
Digital Signals and Systems are everywhere. Digital 1D signals, 2D signals (images) and 3D signals (video) encompass the vast majority of digital information nowadays in several disciplines: -Digital Media, Social Media (music, images, video), -Biome...
Network theory has very many application areas, where graphs are of primary importance, in e.g.,: -Communication networks -Epidemiology -Systems Biology -Social networks. Social Media (e.g., Twitter, Facebook, Instagram, to name a few) has had a tr...
Digital signals are ubiquitous and have many applications in several disciplines: -Digital Media, Social Media (music, voice signals), -Biomedical Signal Analysis and Diagnosis, -Bioinformatics -Autonomous cars, drones, marine vessels, robots -Big ...
Both fixed wing and multicopter UAVs have had a tremendous growth in many applications, such as: -Media Production -Infrastructure Inspection -Surveillance -Semantic mapping -Agriculture inspection -Transportation of goods.
Autonomous Car research and development is booming today, with many companies developing such cars, e.g., TESLA, BMW, MERCEDES, FIAT, TOYOTA, to name a few. Very many application areas emerge from companies like UBER and LYFT, that tend to change the cit...
-Wireless Communication Networks that have many applications in autonomous systems - Cryptography that has many applications in Communications and Blockchain
3D images are an integral part of scientific imaging, used, e.g., in: -Biomedical imaging (e.g., CT MRI scans) -3D microscopy -Geophysical Prospecting. Furthermore, special types of 3D imaging, notably stereo, are routinely used in 3D cinema and 3...
3D Computer Graphics and Virtual Reality play a very important role in computer games, TV/media production and in scientific/medical imaging.
There are many useful CVML Development and Programming Tools. -CUDA has many applications in GPU computing and Deep Neural Networks. -SW tools, libraries and environments used in computer vision and machine learning. -GPU and Multicore CPU Architectu...
Nowadays, Artificial Intelligence drives scientific and economic growth worldwide. This is largely due to advances in Machine Learning (ML). Its applications span and revolutionize almost every human activity: - Autonomous Systems (cars, drones, vessels)...
Nowadays, Artificial Intelligence, notably Advanced Machine Learning (ML) drives scientific and economic growth worldwide. They are essentially massive ‘learning by experience/examples’ systems. However, as our tasks and the world change, such systems sho...
Recognizing existing visual concepts from an image or a video
Part of the AI Mellontology Symposium series
Showcasing EU AI PhD Programs & Discussing ChatGPT on Education
Our dream to make machines sense and perceive (notably see) comes true: nowadays Computer Vision enables diverse applications: -Autonomous Systems (cars, drones, vessels) Perception, -Robotics Perception and Control, -Intelligent Human-Machine Intera...
Our dream to make machines sense and perceive (notably see) comes true: nowadays Computer Vision enables diverse applications: -Autonomous Systems (cars, drones, vessels) Perception, -Robotics Perception and Control, -Intelligent Human-Machine Intera...