Drivers' inattention and distraction detection are crucial features for driver monitoring and driver assistance systems (DAS) and a key point to avoid car crashes. Currently available datasets mainly track head pose and only a few were collected inside a car. Therefore, we present a novel public dataset designed to train algorithms that estimate driver gaze zone in real driving conditions. We provide labeled frame-by-frame images from 19 car points. Depth 3D images were also captured and they are aligned with infrared images. This new dataset is the largest one available for this purpose and the only one that provides 2D and 3D data aligned at pixel using a Intel ^® RealSense™ camera. Finally, in order to establish a baseline, we present a gaze zone estimator built using a transfer learning method.

Background: Climate change, driven by human activity, is rapidly changing our environment and posing an increased risk to human health. Local governments must adapt their cities and prepare for increased periods of extreme heat and ensure that marginalized populations do not suffer detrimental health outcomes. Heat warnings traditionally rely on outdoor temperature data which may not reflect indoor temperatures experienced by individuals. Smart thermostats could be a novel and highly scalable data source for heat wave monitoring. Objective: The objective of this study was to explore whether smart thermostats can be used to measure indoor temperature during a heat wave and identify houses experiencing indoor temperatures above 26°C. Methods: We used secondary data—indoor temperature data recorded by ecobee smart thermostats during the Quebec heat waves of 2018 that claimed 66 lives, outdoor temperature data from Environment Canada weather stations, and indoor temperature data from 768 Quebec households. We performed descriptive statistical analyses to compare indoor temperatures differences between air conditioned and non–air conditioned houses in Montreal, Gatineau, and surrounding areas from June 1 to August 31, 2018. Results: There were significant differences in indoor temperature between houses with and without air conditioning on both heat wave and non–heat wave days (P<.001). Households without air conditioning consistently recorded daily temperatures above common indoor temperature standards. High indoor temperatures persisted for an average of 4 hours per day in non–air conditioned houses. Conclusions: Our findings were consistent with current literature on building warming and heat retention during heat waves, which contribute to increased risk of heat-related illnesses. Indoor temperatures can be captured continuously using smart thermostats across a large population. When integrated with local heat health action plans, these data could be used to strengthen existing heat alert response systems and enhance emergency medical service responses.

Whether a word was bawled, whispered, or yelped, captions will typically represent it in the same way. If they are your only way to access what is being said, subjective nuances expressed in the voice will be lost. Since so much of communication is carried by these nuances, we posit that if captions are to be used as an accurate representation of speech, embedding visual representations of paralinguistic qualities into captions could help readers use them to better understand speech beyond its mere textual content. This paper presents a model for processing vocal prosody (its loudness, pitch, and duration) and mapping it into visual dimensions of typography (respectively, font-weight, baseline shift, and letter-spacing), creating a visual representation of these lost vocal subtleties that can be embedded directly into the typographical form of text. An evaluation was carried out where participants were exposed to this speech-modulated typography and asked to match it to its originating audio, presented between similar alternatives. Participants (n=117) were able to correctly identify the original audios with an average accuracy of 65%, with no significant difference when showing them modulations as animated or static text. Additionally, participants' comments showed their mental models of speech-modulated typography varied widely.