The CP-QAE-I: A video dataset for exploring the effects of personality and culture on perceived quality and affect in multimedia

Abstract

Perception of quality and affect are subjective, driven by a complex interplay between system and human factors. Is it, however, possible to model these factors to predict subjective perception? To pursue this question, broader collaboration is needed to sample all aspects of personality, culture, and other human factors. Thus, an appropriate dataset is needed to integrate such efforts. Here, the CP-QAE-I is proposed. This is a video dataset containing 144 video sequences based on 12 short movie clips. These vary by: frame rate; frame dimension; bit-rate; and affect. An evaluation by 76 participants drawn from the United Kingdom, Singapore, India, and China suggests adequate distinction between the video sequences in terms of perceived quality as well as positive and negative affect. Nationality also emerged as a significant predictor, supporting the rationale for further study. By sharing the dataset, this paper aims to promote work modeling human factors in multimedia perception.