Computer Vision News - January 2020

2 Summary Challenge 2 a forum to partake in organized challenges, publish novel work, and collaborate on the topic kept it going now for four consecutive years. Automatic kinship recognition has many use-cases . For instance, as an aid in forensic investigations (e.g., had we known the Boston Bombers were brothers…). To boot, many beneficiaries from such technologies: consumers (e.g., photo library management), scholars (e.g., ancestral andgenealogical studies), data scientists (e.g., social media analysis and generative-based modeling), prosecutors (e.g., cases of missing children and human trafficking), and humanitarians (e.g., reconnect families across refugee camps). Two critical components must be met to allow for transition from research-to- reality: Every month, Computer Vision News reviews a challenge related to our field. If you are interested to know the new methods proposed by the scientific community to solve the challenges, this section is for you. This month we have chosen to look at the 2020 RFIW large-scale visual kinship recognition data challenge (and workshop) series, held in conjunction with IEEE FG Conference. The website of the challenge, with all its related resources, is here . Read below what Joe Robinson, a PhD Candidate in the SMILE Lab at Northeastern University and one of the organizers of RFIW, tells us about this challenge. Nature-based studies revealed kinship (i.e., blood relatives) is detectable from facial cues . This fact motivated researchers in fields of machine vision and multi-media to focus in the kinship domain for the past decade, with increasing attention the more recent years. With the release of our large-scale Families In theWild (FIW) dataset , along with the annual Recognizing FIW (RFIW) data challenge, both the potential and incentive to advance kin-based vision systems were enhanced: enabling data- driven, deep models to be employed in more realistic experimental settings. The RFIW workshop series provides Recognizing Families In the Wild (RFIW) by Joe Robinson Motivation