IMPLICITY: CITY MODELING FROM SATELLITE IMAGES WITH DEEP IMPLICIT OCCUPANCY FIELDS

Our work addresses the predictability of carbon absorption by ecosystems across the globe, particularly in dry regions. We compare 3 different models, including a deep learning model that can learn from past environmental conditions, and show that this helps improve predictions. Still, challenges remain in dry areas due to varying vulnerabilities to drought. As drought conditions intensify globally, it's crucial to understand the varying impacts on ecosystem function.

Natural disasters often damage buildings and threaten lives, especially in areas with limited resources. To help improve emergency response, we created a global dataset called BRIGHT using both optical and radar images to detect building damage in any weather. We tested many artificial intelligence models and showed how well they work in real disaster scenes. This work can guide better tools for future disaster recovery and help save lives faster.

For many safety-related applications such as road management, well-documented avalanches are important. To enlarge the information, webcams may be used. We propose supporting the mapping of avalanches from webcams with a machine learning model that interactively works together with the human. Relying on that model, there is a 90% saving of time compared to the "traditional" mapping. This gives a better base for safety-critical decisions and planning in avalanche-prone mountain regions.