Each person records 3 different sentences for 7 types of emotions, 21 different sentences per person. A mobile phone records video + another mobile phone records audio + microphone sound card records audio simultaneously. The three devices record simultaneously and are time-synchronized. Person IDs are consistent with 87-1

Each person records 100 sentences. A mobile phone records video + another mobile phone records audio + microphone sound card records audio simultaneously. The three devices record simultaneously and are time-synchronized. Person IDs are consistent with 87-1

1 person, 1 front photo + 1 video 2. Front photo 3. Head circle video 15-19 years old: 2221 people, 20-39 years old: 2816 people, 40-59 years old: 1993 people, 60+ years old: 2970 people

100,000 images, each with 1 face, annotated with 57 key points.

Age distribution: 18-24:25-34:35-44:45-54:55-60 = 1:2:2:2:1

I. 5 photos collected per person: resolution 3024 X 4032, front photo, 45 degrees left photo, 90 degrees left photo, 45 degrees right photo, 90 degrees right photo II. 9 videos, resolution 3840 X 2160, 30 FPS: 1. Turn left to 90 degrees, 2. Turn right to 90 degrees, 3. Look up, 4. Look down, 5. Head circle, 6. Close eyes, 7. Blink, 8. Open mouth, 9. Wear mask

Each person records 1 Live Link Face video, reads 21 sentences in the video, and exports video and facial motion tracking files

nan

7 emotions, 10 sentences for each emotion, totaling 70 sentences, 70 videos. One mic records sound, one mobile phone records sound, one mobile phone records video.

7 emotions, 10 sentences for each emotion, totaling 70 sentences, 70 videos recorded using Live Link Face.