Collection (8)

Poster Session 2023 - AI Internships@Gemeente Amsterdam

Iva Gornishka, Hannah Verstoep

Article
- Gemeente Amsterdam Cluster Digitalisering, Innovatie en Informatie,
- University of Amsterdam
- Iva Gornishka
Unsupervised Object Discovery with Urban Context Cues

Sidewalk accessibility is crucial for safe and comfortable pedestrian travel, especially for citizens with reduced mobility. Manual inspection is expensive and time-consuming, so alternative methods like remote crowdsourcing and computer vision have been explored.

This poster presents Andrea Lombardo's pipeline to address this issue using an unsupervised algorithm to localize accessibility features.

This research was conducted by Andrea Lombardo in collaboration with AI Team, Urban Innovation and R&D, City of Amsterdam.

Involved civil servants: Diederik Roijers

Supervisors: Tim Alpherts & Diederik Roijers
Article
- Gemeente Amsterdam Cluster Digitalisering, Innovatie en Informatie
- Academic,
- Methodology,
- Quantitative,
- Model,
- By students
- Nature & Environment
- Iva Gornishka
- Daan Bloembergen
Individual Tree Delineation

This poster presents Jorges Nofulla's approach to accurately identifying and delineating individual tree trunks in a point cloud dataset.

This research was conducted by Jorges Nofulla in collaboration with the department of Research and Statistics (Onderzoek en Statistiek, O&S) and the AI Team, Urban Innovation and R&D, City of Amsterdam.

Involved civil servants: Nico de Graaff & Daan Bloembergen

Supervisors: Sander Oude Elberink, Nico de Graaff & Daan Bloembergen
Article
- Gemeente Amsterdam Cluster Digitalisering, Innovatie en Informatie,
- University of Amsterdam
- Research plan,
- Academic,
- Methodology,
- By students
- Iva Gornishka
- Daan Bloembergen
Point Clouds Anonymisation

Point Clouds have already proven invaluable for numerous applications related to asset management. While the current density of the point clouds used at the City of Amsterdam is not high enough to recognize people, we would like to prepare for future technological developments and ensure in advance that citizens' privacy can be preserved while data can be shared with the public to stimulate further research and valuable applications.

This poster presents Kacper Sawicz's pipeline to pedestrian anonymisation by training a semantic segmentation model on an existing dataset from Paris and fine-tuning it on Amsterdam data.

This research was conducted by Kacper Sawicz in collaboration with the AI Team, Urban Innovation and R&D, City of Amsterdam.

Involved civil servants: Daan Bloembergen

Supervisors: Martin Oswald & Daan Bloembergen
Article
- Gemeente Amsterdam Cluster Digitalisering, Innovatie en Informatie
- Research plan,
- Academic,
- Methodology,
- By students
- Digital,
- Public space
- Iva Gornishka
- Daan Bloembergen
Lidar-based Window Detection in Urban Environments

A 3D model of the city could aid processes such as granting building permits, it would help the municipality to visualize new policies or urban plans, and it would generally help citizens to envision their city and participate in making decisions about it.

This poster presents Willem van der Vliet's approach to improving the existing 3D model of Amsterdam by more accurately representing windows and doors. This will be done using sensor fusion, focusing on LIDAR sensor data.

This research was conducted by Willem van der Vliet in collaboration with the 3D Amsterdam Team and the AI Team, Urban Innovation and R&D, City of Amsterdam.

Involved civil servants: Wietse Balster & Daan Bloembergen

Supervisors: Holger Caesar & Daan Bloembergen
Article
- Gemeente Amsterdam Cluster Digitalisering, Innovatie en Informatie,
- Vrije Universiteit Amsterdam
- Research plan,
- Academic,
- Methodology,
- By students
- Inclusion & Participation,
- Municipal services
- Iva Gornishka
- Daan Bloembergen
Automatic Text Simplification for Low-Resource Domains

The Municipality of Amsterdam is responsible for communicating in simple language to all its citizens. Unfortunately, automatically aiding editors in their writing process relies on models requiring substantial training data. Unfortunately, simplification data is not readily available for Dutch, let alone for the specific domain of governmental communication and underlying topics such as taxes, health, environment, etc.

This poster presents Daniel Vlantis' research on reproducing an existing low-resource solution from the medical domain and adapting it to general municipal communication.

This research was conducted by Daniel Vlantis in collaboration with AI Team, Urban Innovation and R&D, City of Amsterdam.

Involved civil servants: Iva Gornishka

Supervisors: Shuai Wang & Iva Gornishka
Article
- Gemeente Amsterdam Cluster Digitalisering, Innovatie en Informatie,
- Vrije Universiteit Amsterdam
- Research plan,
- Academic,
- Methodology,
- By students
- Iva Gornishka
Automatically Extracting Venue Accessibility Information Noisy Data

One of the biggest challenges that people with disabilities encounter in the city is that whenever they go to a location (a restaurant, a library, or any other public building), the process of determining in advance whether they can reach the place, if they could enter it and whether they would feel comfortable inside is too long and not reliable. Previous research by the City of Amsterdam has shown that it is possible to automatically extract venue accessibility information from online reviews.

This poster presents Mylène Brown-Coleman's research plans to adapt the existing pipeline to work with noisy data.

This research was conducted by Mylène Brown-Coleman in collaboration with AI Team, Urban Innovation and R&D, City of Amsterdam.

Involved civil servants: Iva Gornishka

Supervisors: Xander Wilcke & Iva Gornishka
Article
- Gemeente Amsterdam Cluster Digitalisering, Innovatie en Informatie,
- University of Amsterdam
- Research plan,
- Academic,
- Methodology,
- By students
- Digital,
- Municipal services
- Iva Gornishka
Towards Automatic Answering of City Council Questions

Standard approaches to question answering do not readily apply to answering complex policy-related questions. Furthermore, existing models and resources primarily focus on English and general questions.

This poster presents Natali Peeva's approach to investigating to what extent and how can state-of-the-art natural language processing models aid the answering City Council questions within the City of Amsterdam. The methodology can later be applied to any policy-related questions independent of their origin.

This research was conducted by Natali Peeva in collaboration with AI Team, Urban Innovation and R&D, City of Amsterdam.

Involved civil servants: Iva Gornishka

Supervisors: João Pereira & Iva Gornishka
Article
- Research plan,
- Academic,
- Methodology,
- By students
- Infrastructure,
- Public space
- Iva Gornishka
Predicting Quay Walls Deformation: A Model Fusion Approach

Predicting quay wall deformation is crucial for maintaining the safety, accessibility and livability of our city.

This poster presents Julian El-Fasih's approach to predicting this deformation automatically. Previous research of the City of Amsterdam has shown that various data sources could be helpful for the task. Julian plans to experiment with different ways of fusing information from these data sources.

Credits poster design: https://morrislift.com/

This research was conducted by Julian El-Fasih in collaboration with the Bridges and Quay Walls Program (Programma Bruggen en Kademuren) and the AI Team, Urban Innovation and R&D, City of Amsterdam.

Involved civil servants: Pantelis Karamitopoulos

Supervisors: Frank Nack & Pantelis Karamitopoulos