Koichi Kise

Clouds frequently degrade optical satellite imagery, limiting the reliability of remote sensing models. However, in the literature, cloud-free analyses are often performed by excluding cloudy images from machine learning datasets and methods. This restricts their usefulness in time-critical scenarios such as disaster response, where waiting for cloud-free imagery is impractical. Cloud removal can mitigate this issue, but methods remain imperfect and may introduce visual artifacts. Therefore, it is desirable to develop cloud-robust methods by combining optical imagery with radar data, a modality unaffected by clouds. While datasets for machine learning combine optical and radar data, most researchers exclude cloudy images from training and evaluation. We identify this exclusion as a limitation that reduces applicability to cloudy scenarios and address it by introducing CloudyBigEarthNet (CBEN), a dataset of paired optical and radar images containing cloud occlusions for land-use and land-cover classification. Using average precision (AP), we show that state-of-the-art methods trained on clear-sky optical and radar data suffer performance drops of between 23.8 and 33.4 AP points when tested on cloudy imagery. We adapt these methods using cloudy images during training and improve AP on cloudy test cases by 17.2 to 28.7 AP points. Code and dataset have been published.

Commercially available wearable health devices (WHDs) carry the potential to decentralize healthcare systems. These devices can empower individuals with health knowledge by offering a low-cost and accessible way to monitor physical activity, sedentary behavior, cardiac health, and sleep. However, a lack of standardization in design, health, and safety regulations means that consumer-grade WHDs on the market vary in efficacy to affect positive behavior change in users, as user compliance alone does not indicate whether these devices actually influence wellbeing outcomes long term. We use a grounded theory analysis of the experiences of seven long-term informed users of the same wearable, the Oura Ring, to propose a substantive theory describing the tacit challenges that these users face in order to truly benefit from their device even after extended use. We provide recommendations as to how designers of wearable devices can facilitate the user’s journey to surpass these obstacles.

Self-tracking technologies and wearables automate the process of data collection and insight generation with the support of artificial intelligence systems, with many emerging studies exploring ways to evolve these features further through large-language models (LLMs). This is done with the intent to reduce capture burden and the cognitive stress of health-based decision making, but studies neglect to consider how automation has stymied the agency and independent reflection of users of self-tracking interventions. In this position paper, we explore the consequences of automation in self-tracking by relating it to our experiences with investigating the Oura Ring, a sleep wearable, and navigate potential remedies.