Proceedings of the 9th International on Audio/Visual Emotion Challenge and Workshop

Abstract

It is our great pleasure to welcome you to the 9th Audio-Visual Emotion Challenge - AVEC'19, held in conjunction with the ACM Multimedia 2019 in Nice, France. This year's challenge and associated workshop continues to push the boundaries of audio-visual emotion and health recognition towards real-life applications of affective computing. Looking back in the history of AVEC, the first challenge posed the problem of detecting discrete emotion classes on a large set of natural behaviour data. The second AVEC extended this problem to the prediction of continuous valued dimensional affect, which was enlarged for the third edition to include the prediction of self-reported severity of depression, and with enriched annotations for the fourth edition. Physiological signals were then introduced for the prediction of dimensional affect in the fifth AVEC. In the sixth edition, the focus was on depression analysis from human-agent interactions in a Wizardof- Oz paradigm, and on emotion recognition from human behaviours captured 'in-the-wild' for the seventh AVEC. Cross-cultural affect prediction 'in-the-wild' was then introduced in the eighth edition, along with classification of bipolar disorder. This year, we pursue the road of cross-cultural affect recognition in real-life conditions, this time with both Western-European and Chinese cultures as tested population. The automatic detection of human state-of-mind, which is pivotal for mood related disorders, and the level of depression from interviews conducted by a virtual agent wholly driven by artificial intelligence, are further introduced in AVEC 2019. The main underlying motivation of the AVEC series is the need to advance emotion and health estimation for multimedia retrieval to a level where behaviours can be reliably sensed in real-life conditions, as this is precisely the type of data that the new generation of affect-oriented multimedia and human-machine/human-robot communication interfaces have to face in the real world.