Assets: 3,000 participants
25 - 50 images per participant (75,000 - 150,000 images)
1 video per participant (3,000 videos total)
Description of Deliverables
• Type of asset: Images and Videos
• Duration of Videos assets : 20 to 40 seconds
• Will the Asset contain PII? Yes
• Capture SW/HW : Any device that can satisfies the following minimum requirement will be acceptable :
3MP images [HEIC, JPEG, PNG]
720P videos [ MP4, Mov]
• Participants demographic – Participants should be evenly distributed between the following groups :
Vision Impairment- Significant Vision loss or blindness
Hearing Impairment- Deafness
Motor Impairment- Tremors (in Head/ neck), Low dexterity( in head/ neck), Limited facial control(such as after stroke, Facial Tricks ), Limited head movement/ control.
Cognitive Impairment- Autism Spectrum, Alzheimer’s Disease
Facial /Head diversity –Large Facial Birthmarks, burns , scarring, Pigmentation differences, clef lip palate , Hemifacial microsomia, other facial/ cranial conditions
Language/speech Impairment- Stutter, Articulation impairment, Lisp or other Speech impediment, Other voice Disorder.
• Collection should NOT take place in the following countries: China, EU Member States, Argentina, Japan, Malaysia, Mexico, South Korea, Brazil, UAE, UK, Canada, Australia, New Zealand, Colombia, Costa Rica, Hong Kong, Macau, Israel, Morocco, Peru, Philippines, Russia, Singapore, South Africa, Taiwan, Tunisia, Turkey, and Uruguay.
Specifications and Protocol
• Image Requirements –
Images should be from participants photos Library.
Volume-
Target # of Images: 25 – 50
o Minimum of 25 images per participant
o Maximum of 50 images per participant
Image Date/timeframe- Any photo from past 10 years
As much as possible, image sets should include variation in:-
Hairstyles
Accessories
Facial expression
Lighting conditions
Backgrounds/environments (i.e. indoor/outdoor photos)
Size of person in the frame, i.e. headshot/selfie, medium-shots (upper body) and long-shots (fully body)
Each image should predominantly show the contributor in the frame, with at least 50% of their face visible (not necessarily looking directly at the camera, but visible - a profile face is acceptable)
Additional Image Requirements:
Face clearly visible in all photos and high diversity in head pose (frontal, 3/4, profile, looking up, looking down, etc.), expression, lighting conditions, clothes.
Diversity in headwear, glasses, facial hair, make-up is a bonus.
Other persons may be shown in the photos only if the contributor has the authority to submit and consent to the use of the images of the other persons
Background: any, no specific distribution
Lighting Conditions: any, no specific distribution
Camera Parameters: any, no specific distribution
Distance to Subject: Head size should be a minimum of 200 x 200px in original resolution.
• Video Requirements:
1 videos per contributor
Length: 20-40 seconds
Movements should be slow and deliberate to reduce the risk of motion blur
Background: neutral, solid background
Video Protocol:-
The entire face should be visible throughout the selfie video, with the following actions:
o Cross (look up, down, right and left.)
o Full circle (Head should roll in a clockwise or counter-clockwise motion)
o Short sentence, “Go <Insert Vendor Name here”
o Smile
o Two blinks
o V sign (peace sign)
Lighting: Indoor overhead lighting
No face/head accessories should be worn other than daily worn accessories (e.g. eye glasses, hearing aids, turbans, etc.)
o For example, participants should not wear large hair bows or hats unless those are worn daily or frequently