5
Born 1989, photos from 2015-2018.
I tried using Google's Cloud Vision API to estimate the nasal index for 8 of her photos: https://cloud.google.com/vision/docs/drag-and-drop.
The API doesn't provide a landmark for the nasion, but it does include a landmark for the point between the eyes, which I is approximately equal at least in this case (even though for some people, the nasion is much higher on the y-axis or z-axis than the middle point between the eyes.)
Based on the measurements returned by the API, the nasal index ranged from about 101 to 107 depending on the photo:
$ gcl(){ curl -HAuthorization:Bearer\ `gcloud auth application-default print-access-token` -HContent-Type:application/json\;charset=utf-8 https://vision.googleapis.com/v1/"$1" "${@:2}";}
$ gface(){ gcl images:annotate -d@<(printf %s "$(for x;do printf %s '{"image":{"content":"'"`base64 "$x"`"'"},"features":[{"maxResults":1,"type":"FACE_DETECTION"}]},';done|sed '1s/^/{"requests":[/;$s/.$/]}/')");}
$ faced(){ jq --arg x "$1" --arg y "$2" -r '.responses[].faceAnnotations[].landmarks[]|select(.type==$x or.type==$y)|.position|[.x,.y,.z]|map(tostring)|join(" ")' "${@:3}"|paste - -|awk '{print(($1+$4)^2+($2+$5)^2+($3+$6)^2)^.5}';}
$ gface files/*.jpg>output.json
$ paste <(faced NOSE_BOTTOM_LEFT NOSE_BOTTOM_RIGHT output.json) <(faced MIDPOINT_BETWEEN_EYES NOSE_BOTTOM_CENTER output.json)|awk '{print 100*$1/$2"\t"$0}'|tr \\t \;
100.846;726.952;720.851
106.585;748.674;702.421
105.948;939.712;886.954
102.056;1357.4;1330.05
100.872;1256.48;1245.62
102.056;1357.4;1330.05
100.765;1801.27;1787.59
101.919;1985.73;1948.35
According to a definition of morphological nasal index by Eickstedt (1962), platyrrhine is between 85 and 100 and hyperplatyrrhine is above 100 (http://humanphenotypes.net/metrics/nasalindex.html).
Bookmarks