Mercan Topkara

Mercan Topkara, Thomas Erickson, Umut Topkara et al.

Quick response codes - 2D patterns that can be scanned to access online resources - are being used in a variety of industrial and consumer applications. However, it is problematic to use multiple QR codes in close proximity: scans can fail or result in access to the wrong resource. While this problem is, strictly speaking, due to the design of the scanning software, the very large number of extant scanning applications makes changing the software a difficult logistical challenge. Instead, we describe the design of a new type of QR code that not only enables the use of multiple QR codes in close proximity, but also is compatible with existing scanning solutions. In an evaluation with 20 users, it was found that the new QR codes were as usable as traditional ones, and that they were superior for selecting one code from many. Users did have initial difficulty in discovering how to use the new QR code, so further work is required on that front. We conclude with a discussion of the pros and cons of pQR codes.

Brendan Jou, Tao Chen, Nikolaos Pappas et al.

Every culture and language is unique. Our work expressly focuses on the uniqueness of culture and language in relation to human affect, specifically sentiment and emotion semantics, and how they manifest in social multimedia. We develop sets of sentiment- and emotion-polarized visual concepts by adapting semantic structures called adjective-noun pairs, originally introduced by Borth et al. (2013), but in a multilingual context. We propose a new language-dependent method for automatic discovery of these adjective-noun constructs. We show how this pipeline can be applied on a social multimedia platform for the creation of a large-scale multilingual visual sentiment concept ontology (MVSO). Unlike the flat structure in Borth et al. (2013), our unified ontology is organized hierarchically by multilingual clusters of visually detectable nouns and subclusters of emotionally biased versions of these nouns. In addition, we present an image-based prediction task to show how generalizable language-specific models are in a multilingual context. A new, publicly available dataset of >15.6K sentiment-biased visual concepts across 12 languages with language-specific detector banks, >7.36M images and their metadata is also released.