I'm Pietro De Angeli

Reimplemented the paper Efficient Temporally-Aware DeepFake Detection using H.264 Motion Vectors. Matched the reported accuracy (71%) while reducing inference cost by replacing MTCNN with a lightweight YuNet-based face detector.

Implemented the paper ”Layered Vectorization via Semantic Simplification” introducing radial gradient shades to improve the vectorization quality of images with smooth color transitions. It hasn't reached MSE improvment but it better represent shades . Repo is not public yet

Developed a fake news detection classifier (logistic classification and Neural network) with OpenAI text embeddings. Reached an accuracy of 99%.

Work in Progress. Trying to improve the results of Max-Cut problem using evolutionary algorithms. Implemented a metaheuristic algorithm based on the Inspyred library to optimize the parameters of a classical heuristic algorithm for Max-Cut problem.