Proposte Tesi
SERVIZI PER LA DIDATTICA
Home
Created with Raphaël 2.1.0
Servizi per gli studenti
Created with Raphaël 2.1.0
Proposte Tesi

Ricerca CERCA
  KEYWORD

Computer Vision Methods for Microscopy Imaging

azienda Tesi esterna in azienda    


Parole chiave AI, COMPUTER VISION, DEEP LEARNING, MACHINE LEARNING

Riferimenti RAFFAELLO CAMORIANO

Riferimenti esterni Riccardo Volpi, Ph. D., Matteo Zanotto, Ph. D.

Gruppi di ricerca Arsenale Bioyards, DAUIN - GR-23 - VANDAL - Visual and Multimodal Applied Learning Lab

Tipo tesi EXPERIMENTAL, RESEARCH AND DEVELOPMENT, START UP

Descrizione Arsenale Bioyards is aiming to make biomanufacturing via precision fermentation economically viable at industrial scale – for the very first time. We are looking for a motivated student to carry out a six month project with us, in which s/he will develop novel solutions to monitor microorganisms cultures using microscopy images and computer vision techniques, such image segmentation, detection and classification. The methods will be developed in Python and the student will have access to high-performance GPUs. The student will have the chance to work in a vibrant startup environment, and to develop impactful solutions to important problems.

Location: Pordenone / Work From Home friendly.
Duration: Six months
Note: this is a paid thesis internship
Contact: raffaello.camoriano@polito.it

Goals
● Literature review, focusing on both microscopy imaging and the application of computer vision methods to this data modality
● Formalizing our problem
● Helping in the data collection and its annotations
● Implementing state-of-the-art methods from the literature
● Testing on public benchmark and on data collected in our Lab
● (Optional) Designing and implementing solutions that address limitations found in existing methods from the literature
● (Optional) Depending on the outcome of the project, we will write a research paper about it

References
● Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for ImageRecognition at Scale, ICLR 2021
● Sypetkowski et al., RxRx1: A Dataset for Evaluating Experimental Batch Correction Methods, CVPR Workshops 2023
● Kraus et al., Masked Autoencoders for Microscopy are Scalable Learners of Cellular Biology, CVPR 2024
Requirements

Conoscenze richieste Requirements
● Motivated to work in a dynamic, fast-paced environment
● Strong Python skills
● Knowledge of Pytorch
● Has worked on machine learning and computer vision projects


Scadenza validita proposta 29/01/2026      PROPONI LA TUA CANDIDATURA