Domain adaptation per classificazione di point cloud mediante pseudo-annotazioni

La classificazione di dati geometrici 3D come point cloud è un tema emergente nell'ambito della visione artificiale in quanto trova applicazione in molteplici contesti di guida autonoma, robotica e realtà aumentata. Sebbene nel mercato siano presenti una grande quantità di sensori in grado di ottenere scansioni reali, la loro annotazione costituisce un collo di bottiglia per la generazione di dataset. Per sopperire al problema si ricorre spesso alla domain adaptation sfruttando dati sintetici annotati. Questo elaborato si pone come obiettivo l'analisi e l'implementazione di metodi di domain adaptation per classificazione di point cloud mediante pseudo-labels. In particolare, sono stati condotti esperimenti all'interno del framework RefRec valutando la possibilità di sostituire nuove architetture di deep learning al modello preesistente. Tra queste, Transformer con mascheramento dell'input ha raggiunto risultati superiori allo stato dell'arte nell'adattamento da dati sintetici a reali (ModelNet->ScanNet) esaminato in questa tesi.

Deep learning based style transfer for low altitude aerial imagery

Unmanned Aerial Vehicle (UAVs) equipped with cameras have been fast deployed to a wide range of applications, such as smart cities, agriculture or search and rescue applications. Even though UAV datasets exist, the amount of open and quality UAV datasets is limited. So far, we want to overcome this lack of high quality annotation data by developing a simulation framework for a parametric generation of synthetic data. The framework accepts input via a serializable format. The input specifies which environment preset is used, the objects to be placed in the environment along with their position and orientation as well as additional information such as object color and size. The result is an environment that is able to produce UAV typical data: RGB image from the UAVs camera, altitude, roll, pitch and yawn of the UAV. Beyond the image generation process, we improve the resulting image data photorealism by using Synthetic-To-Real transfer learning methods. Transfer learning focuses on storing knowledge gained while solving one problem and applying it to a different - although related - problem. This approach has been widely researched in other affine fields and results demonstrate it to be an interesing area to investigate. Since simulated images are easy to create and synthetic-to-real translation has shown good quality results, we are able to generate pseudo-realistic images. Furthermore, object labels are inherently given, so we are capable of extending the already existing UAV datasets with realistic quality images and high resolution meta-data. During the development of this thesis we have been able to produce a result of 68.4% on UAVid. This can be considered a new state-of-art result on this dataset.

Hierarchical Multi-Label Text Classification in a Low-Resource Setting

In this thesis we address a multi-label hierarchical text classification problem in a low-resource setting and explore different approaches to identify the best one for our case. The goal is to train a model that classifies English school exercises according to a hierarchical taxonomy with few labeled data. The experiments made in this work employ different machine learning models and text representation techniques: CatBoost with tf-idf features, classifiers based on pre-trained models (mBERT, LASER), and SetFit, a framework for few-shot text classification. SetFit proved to be the most promising approach, achieving better performance when during training only a few labeled examples per class are available. However, this thesis does not consider all the hierarchical taxonomy, but only the first two levels: to address classification with the classes at the third level further experiments should be carried out, exploring methods for zero-shot text classification, data augmentation, and strategies to exploit the hierarchical structure of the taxonomy during training.