In this paper, we present the dataset Right Inflight developed to support the exploration of the match between video content and the situation in which that content is watched. Specifically, we look at videos that are suitable to be watched on an airplane, where the main assumption is that that viewers watch movies with the intent of relaxing themselves and letting time pass quickly, despite the inconvenience and discomfort of flight. The aim of the dataset is to support the development of recommender systems, as well as computer vision and multimedia retrieval algorithms capable of automatically predicting which videos are suitable for inflight consumption. Our ultimate goal is to promote a deeper understanding of how people experience video content, and of how technology can support people in finding or selecting video content that supports them in regulating their internal states in certain situations. Right Inflight consists of 318 human-annotated movies, for which we provide links to trailers, a set of pre-computed low-level visual, audio and text features as well as user ratings. The annotation was performed by crowdsourcing workers, who were asked to judge the appropriateness of movies for inflight consumption.@en

This paper provides an overview of the JointContest on Multimedia Challenges Beyond Visual Analysis.We organized an academic competition that focused on fourproblems that require e‚ective processing of multimodalinformation in order to be solved. Two tracks were devoted togesture spotting and recognition from RGB-D video, two fundamentalproblems for human computer interaction. Anothertrack was devoted to a second round of the €rst impressionschallenge of which the goal was to develop methods torecognize personality traits from short video clips. For thissecond round we adopted a novel collaborative-competitive(i.e., coopetition) setting. ‡e fourth track was dedicated tothe problem of video recommendation for improving userexperience. ‡e challenge was open for about 45 days, andreceived outstanding participation: almost 200 participantsregistered to the contest, and 20 teams sent predictions inthe €nal stage. ‡e main goals of the challenge were ful€lled:the state of the art was advanced considerably in the fourtracks, with novel solutions to the proposed problems (mostlyrelying on deep learning). However, further research is stillrequired. ‡e data of the four tracks will be available to allowresearchers to keep making progress in the four tracks@en

In this paper we present an overview of the Context of ExperienceTask: recommending videos suiting a watching situationwhich is part of the MediaEval 2016 Benchmark. Theaim of the task is to explore multimedia content that iswatched under a certain situation. The scope of the thisyears task lies on movies watched during a flight. We hypothesizethat users will have different preferences for moviesthat are watched during a flight compared to when a movieis watched at home or the cinema. This is most probablyinfluenced by the context and the devices used to watch. Inthe case of being on a flight, the context is clearly differentto normal situation (noise, compact, bad air) and also thedevices differ (small screens, bad audio quality). The maingoal of the task is to estimate if a person would like to watcha certain movie on the airplane or not. As dataset we providea large collection of movies, collected from an airline,including pre-extracted visual, text and audio features.@en

Crowdsourcing has established itself as a powerful tool for multimedia researchers, and is commonly used to collect human input for various purposes. It is also a fairly widespread practice to control the contributions of users based on the quality of their input. This paper points to the fact that applying this practice in subjective assessment tasks may lead to an undesired, negative outcome. We present a crowdsourcing experiment and a discussion of the ways in which control in crowdsourcing studies can lead to a phenomenon akin to a self-fulfilling prophecy. This paper is intended to trigger discussion and lead to more deeply reflective crowdsourcing practices in the multimedia context.@en