Multimodal Methods for the Analysis and Generation of Images - 2026_CDR_DEIB_4

Scientific-Disciplinary Group

09/IINF-05 - Information Processing Systems

Description

The research program investigates advanced methodologies for image analysis and image generation based on deep learning. The aim is to study techniques for handling multimodal data by integrating visual information (2D and 3D) with textual or tabular metadata. This integration supports tasks such as predictive modeling, anomaly detection, and synthetic data generation. The models developed are expected to exploit metadata to guide and condition image analysis outputs. By leveraging metadata-driven conditioning, models are expected to improve the quality and relevance of synthetic data, strengthen accuracy and robustness in anomaly detection, and enhance outcome prediction in analytical scenarios. The research aims to advance understanding of how multimodal deep learning architectures can combine heterogeneous data sources to enable more adaptive, interpretable, and efficient decision-making across diverse applications in computer vision and data analysis.

Job posting website

https://www.polimi.it/en/bandi-per-contratti-di-ricerca

Number of positions

1

Funding body

Politecnico di Milano

How to apply

https://aunicalogin.polimi.it/aunicalogin/getservizio.xml?id_servizio=1079

Selection process