One more year the ESA Φ-WEEK brings us the latest state of Earth Observation.
One more year the ESA Φ-WEEK brings us the latest state of Earth Observation. This will be the fourth Φ-WEEK and our second time attending the event (although the team has been present since the first edition). This year's focus is on the New Space economy and innovations in Earth Observation.
The conference started with important announcements by ESA's director, Josef Aschbacher, outlining the main challenges that Europe faces nowadays. Among them, climate change and its caused effects (such as floods and fires) are among the top priorities. The use of Sentinel-1 and Sentinel-2 imagery can help us evaluate the impact of this events and assess the resilience of infrastructures, cities and other important assets. his is the main focus of our products, Impact & Vulnerability Pulse. Learn more here.
Another important topic highlighted during the presentation keynote is the commercialisation of Space, with the presentation of a 2025 Agenda based on speed, capital and talent in order to improve the commercial Earth Observation sector. To that end, the creation of a network of Φ-labs excites us the most. Φ-labs are research spaces dedicated to the advancement of the field of Earth Observation with new and exciting technology. We are fully committed to pushing for a Φ-lab Barcelona!
Lastly, a final bet on in-space edge-computing was placed with up to 100k€ invested to the top 10 ideas in this field. Bringing cloud-computing capabilities to Space will advance the field massively. Learn more here. In this line, Φ-sat-2 will bring "APPs" to space, opening on-board sensors and compute to developers very much like we can do nowadays with smartphones. Learn more here.
In last year's Φ-WEEK we presented our startup and our vision for unlocking the value of EO data in order to build a Digital Twin Earth Earth based on Artificial Intelligence where anyone can connect to and extract useful information from EO data without expertise. This year, EarthPulse will be present in the event in the following talks:
📅 Thursday 14th October 17:30 CEST
In this side event the winners of the Enhanced Sentinel 2 Agriculture challenge will be presented and awarded. The new challenges and future challenges will also be presented. If you want to participate in AI4EO challenges, visit platform.ai4eo.eu.
📅 Friday 15th October 09:30 CEST
Most methods and applications in Earth Observation require some type of reference data either for validation or calibration. Machine Learning (ML) is widely used in EO and new ML and Artificial Intelligence (AI) approaches have strong potential to improve information retrievals, processing efficiencies and enable new types of applications in EO.
Unfortunately, the limited availability of quality training and reference datasets has become a major bottleneck limiting more widespread innovation. Moreover, there is a lack of open tools, agreed procedures and best practices for working with training and reference datasets.
This side event brings to together several community initiatives and experts for working with training and reference data in EO. In addition, the event aims to facilitate a community dialogue and discussion (using live sli.do polls etc.) and provides an opportunity to align various ongoing and planned efforts.
At this event, we are going to present the challenges that we face when dealing with training data in AI4EO and how we approach the problem.
Our vision about DL4EO shared in the Space & AI Conference and the AI & Big Data Congress.
We will talk about the current state, barriers and future directions of Deep Learning for EO.
Help us develop the use of Copernicus Sentinel Imagery for land applications!