Amsterdam-Internships's repositories
Self-Learning-Traffic-Lights
Master thesis Artificial Intelligence at the University of Amsterdam as intern at the municipality of Amsterdam.
Pole-Part-Segmentation
This repository contains the implementation of Kernel Point Convolution (KPConv) to decompose the pole-like objects to their constituent part (part segmentation of the pole-like objects).
FewShotCrowdCounting
Few shot learning in the context of cross-scene crowd counting
Venue-Accessibility-Google-Reviews
NLP pipeline for extracting insights related to venue accessibility in Amsterdam.
Point_Cloud_Multi-Scale_Object-Based_Augmentation
MSc Artificial Intelligence Thesis at the UvA in collaboration with the City of Amsterdam. The project concerns the improvement of semantic segmentation quality of underrepresented objects in imbalanced urban point clouds.
Road-Condition-Monitoring
Master Thesis DS with Roads and Mobility Department
Double-Listing
Master Data Science UvA to find double listings in short stays data set
fence-detection
Research internship with the Gemeente Amsterdam concerning the detection of fencing along open water within the city center.
Micromobility-Parking
MSc Data Science (UvA) thesis project for analysing the parking behavior of micromobility vehicles for the assessment of sidewalk accessibility
OldVersion-Venue-Accessibility-NLP
Master Data Science (University of Amsterdam) Thesis Internship project for the Gemeente Amsterdam (Amsterdam for All) Accessibility Project.
sidewalk-accessibility-features
MSc AI thesis project (UvA). Unsupervised Open-World Object Discovery of sidewalks accessibility features in the city of Amsterdam.
super-weak
MSc thesis William Bos. Considers the weak supervision of Dutch corpora for the purpose of creating more training data for fine-tuning PLMs on Dutch NER.
Tree-trunk-segmentation
Individual Tree Trunks Segmentation in a Lidar Point Cloud.
3DBAG_BGT_Intersect
This repository computes the optimial intersection height between 3DBAG building and 3Dfied BGT buildings based on point clouds. Supported output formats are .obj, .ply or .cityjson
Multi-objective_Optimization_using_Interactive_User_Approach
Multi-Objective Optimization problem aims at producing optimal paths for citizens with mobility issues in the city of Amsterdam. This problem will further be developed with user interaction by asking feedbacks from the users and performing preference elicitation strategies on the feedback.
Open3D-ML
An extension of Open3D to address 3D Machine Learning tasks; Forked to adapt and use for Anonymisation of Point Clouds
tree_root_model
A model that determines the current and future required space for tree roots.
Twitter_Feature_Extraction
This repo contains a novel architecture to recognize trends in tweets to extract features that are able to perform classification tasks.