Airbnb is changing the landscape of the hospitality industry. For guests, the process of selecting a place to stay is a type of zero-acquaintance situation, and to this day little is known about the inferences that guests make about Airbnb listings. Environmental psychologists were among the first studying home environments, but most works were based on in situ visits of personal spaces, which makes the process difficult to scale. Our work constitutes a first attempt at understanding how potential Airbnb guests form first impressions from images, one of the main modalities featured on the platform. We collected Airbnb images, focusing on the countries of Switzerland and Mexico as case studies. We then used crowdsourcing mechanisms to gather annotations on physical and ambiance attributes, finding that agreement among raters was high for most of the attributes. We performed a cluster analysis and showed that both physical and psychological attributes could be grouped into three clusters. We then extracted state-of-the-art features from the images to automatically infer the annotated variables in a regression task. Results show the feasibility of inferring ambiance impressions of homes on Airbnb, with up to 42% of the variance explained by our model. In terms of image representations, best results were obtained using activation layers of deep convolutional neural networks trained on the Places dataset, a collection of scene-centric images.