Mitchell O’Hara-Wild @mitchoharawild Monash University" /> class: center, middle, inverse, title-slide # Taipan ## Woman faces machine ### Stephanie Kobakian <span class="citation">@srkobakian</span> <a href="https://twitter.com/srkobakian"><i class="fab fa-twitter "></i></a> </br> Mitchell O’Hara-Wild <span class="citation">@mitchoharawild</span> <a href="https://twitter.com/mitchoharawild"><i class="fab fa-twitter "></i></a> <br> Monash University --- background-image: url(figures/taipan.png) background-position: 90% 50% background-size:50% class: middle ## T<span style="color:orange">ool for</span> ## A<span style="color:orange">nnotating</span> ## I<span style="color:orange">mages in</span> ## P<span style="color:orange">reparation for</span> ## AN<span style="color:orange">alysis</span> --- class: inverse, middle # Motivation - ###Contrast four emotion recognition APIs ## Gold standard: - ###A manual set of classified faces --- background-image: url(figures/2016_HSA_R01_BTomic_AUS_vs_DIstomin_UZB_MS157_clip.0013bg.png) --- background-image: url(figures/2016_HSA_R01_BTomic_AUS_vs_DIstomin_UZB_MS157_clip.0013.png?token=AS7upn8XF4JnVGwfMHBdLb0HS3TRu8goks5bTxK4wA%3D%3D) --- class: center background-image: url(figures/2016_CT6_R01_EKulichkova_RUS_vs_APetkovic_GER_WS128_clip.0080.png) --- class: inverse, middle # Manual Classification Program - ### Shiny app - ### presented two question sets - ### produced one tidy data set --- class: inverse, middle #Shiny > ## A framework for R that enables you to build interactive web applications ### Interactive ### Web based --- background-image: url(figures/mrp1.jpg) background-position: 50% 50% background-size:70% --- background-image: url(figures/mrp2.jpg) background-position: 50% 50% background-size:70% --- background-image: url(figures/mrp2.jpg) background-position: 50% 50% background-size:70% --- # Results snapshot <!-- --> --- background-image: url(figures/2016_CT6_R02_ABeck_GER_vs_TBacsinszky_SUI_WS220_clip.0017.png) # Recognise quirks of each API --- background-image: url(figures/2016_HSA_R03_FDelbonis_ARG_vs_GSimon_FRA_MS302_clip.0072.png) # Net faces --- background-image: url(figures/2016_CT6_R01_CGarcia_FRA_vs_BStrycova_CZE_WS145_clip.0015.png) --- background-image: url(figures/desert.jpg) class: inverse, top # Where to from here? --- # Current survey tools   --- class: inverse, middle # Taipan is a tool for you ## Distributable Shiny surveys ## Easily exported tidy data --- background-image: url(figures/taipan.png) background-position: 90% 50% background-size:50% class: inverse, middle ## Requires - ### Images - ### Questions - ### App Location --- background-image: url(figures/tp1.jpg) background-size:100% --- background-image: url(figures/tp2.jpg) background-size:100% --- background-image: url(figures/tp3.jpg) background-size:100% --- background-image: url(figures/tp4.jpg) background-size:100% --- background-image: url(figures/tp5.jpg) background-size:100% --- background-image: url(figures/tp6.jpg) background-size:100% --- # Installation ```r devtools::install_github("srkobakian/taipan", force = TRUE) ``` -- ## Usage ```r library(taipan) buildTaipan( questions = taipanQuestions(questions), "images", file.path(here::here(), "taipan_app"), overwrite = TRUE) ``` --- class: inverse, middle, centre #Thank you! ## useR! 2018 ## Stephanie Kovalchik Tennis Australia, Victoria University ## Dianne Cook Monash University --- background-image: url(figures/taipan.png) background-position: 50% 5% background-size: 40% class: bottom, centre, inverse ### Slides created via xaringan ### https://github.com/srkobakian/taipan ### This work is licensed under [<i class="fab fa-creative-commons "></i> BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/)