ANC Workshop - Oisin Mac Aodha and Seth Sohan

 

Speaker: Oisin Mac Aodha

 

Title: Computer Vision for Expert Derived Data

 

Abstract:

Expert tasks are domain-specific tasks which require uncommon knowledge or skill, e.g. diagnosing pathologies in medical image data or identifying different species of plants and animals. Tasks of this form often embody many interesting properties, e.g. they can contain fine-grained visual concepts, supervision can be limited and often imperfect, they can exhibit interesting taxonomic structure, and have non-visual side information. In this talk, I will discuss some of our recent work in this space, spanning self-supervised representation learning, weakly supervised object localization, and approaches for modelling the spatio-temporal distributions of objects.

 

Speaker: Seth Sohan

 

Title: Census-Independent Population Estimation using Representation Learning

 

Abstract:

Knowledge of population distribution is critical for building infrastructure, distributing resources, and monitoring the progress of sustainable development goals. Although censuses can provide this information, they are typically conducted every ten years with some countries having forgone the process for several decades. Population can change in the intercensal period due to rapid migration, development, urbanisation, natural disasters, and conflicts. Census-independent population estimation approaches using alternative data sources, such as satellite imagery, have shown promise in providing frequent and reliable population estimates locally. Existing approaches, however, require significant human supervision, for example annotating buildings and accessing various public datasets, and therefore, are not easily reproducible. We explore recent representation learning approaches, and assess the transferability of representations to population estimation in Mozambique. Using representation learning reduces required human supervision, since features are extracted automatically, making the process of population estimation more sustainable and likely to be transferable to other regions or countries. We compare the resulting population estimates to existing population products from GRID3, Facebook (HRSL) and WorldPop. We observe that our approach matches the most accurate of these maps, and is interpretable in the sense that it recognises built-up areas to be an informative indicator of population.

 

https://arxiv.org/abs/2110.02839

 

 

 

 

 

 

 

 

 

 

 

 

Mar 22 2022 -

ANC Workshop - Oisin Mac Aodha and Seth Sohan

Tuesday, 22nd March 2022

online