Our idea consists in creating an AI model to analytically estimate a perceived safety score for every image being analysed:
Firstly, we are asking you to help us locate unsafe hotspots or places needing intervention (so that we can validate our previous model), while we start working with urban labs on solutions to promote change. If you can, upload a photo of the place and leave a comment to caption it. You may also vote on the pins already placed - we are trying to empower cyclists in fighting for their own safety, so it is all about community values. Let's make our home a safer, greener, cleaner place. Next steps will be further announced.
Thank you!
Miguel Peliteiro, a tech-savyy MD, is run over by a car while cycling on a bike path. Critically injured, induced coma, 5 surgeries, 4-month hospital stay. Currently thriving to start medical practice.
Luís Rita, a skillful data scientist and daily bike user, develops Become a Better Cyclist with Deep Learning at Imperial College London. Currently working on his PhD there. Top Talent Under 25 in the meanwhile.
The two linked, brainstormed and decided to take action. Ended up drafting this ambitious project. CycleAI is born.
CycleAI is the winner of the European Cyclists' Federation Hackathon. And has now chances of being featured in the United Nations website and selected as one of the most innovative projects. Pitching in the world’s biggest cycling conference: Velo-City 2021 Lisbon.
CycleAI has a new member. Known for its entrepreneurial projects and a singular mind in Machine Learning and Web Development, Codrin Bostan embarks on this adventure.
With the technical support from Amazon and a prize money to accelerate the project, we now have the tools to present the first results.
With Building Global Innovators and Área Metropolitana de Lisboa, we will implement our project in the capital. Its urban area comprises 3M people out of the 11M in Portugal.
Finished a MSc in Computing from Imperial College London, now embarks in CycleAI as our Software Developer and responsible for the Machine Learning component.
First results from our poll are out – crowdsourcing phase 1 is complete. Onto other crowdsourcing models. Coding on the way. Time to start changing urban planning!