Research Article

2.5D Facial Personality Prediction Based on Deep Learning

Table 1

Summary of the literatures on the accuracy of personality prediction.

ReferencesN photoAssessmentTraitHighest accuracy (%)

[12, 13]186Self-reportConscientiousness (M)81.56
Conscientiousness (F)82.22
Skepticism (M)72.64
Skepticism (F)82.22
[14, 15]66Other reportDominance91.23
Threatening90
Extraversion
[7]244Self-reportOpenness, striving, and domination (F)63
Reliability, friendliness, and responsibility (M)65
[16]12,447Self-reportBig Five (BF)58
[17]608Self-reportNeuroticism82.35
Extraversion84.31
Rigorism84.31
[18]829Other reportNeuroticism, openness, and extraversion65
[19]1856Other reportCriminality89.51
[15]220Other reportWarmth76
Reliability81
[20]480Other reportIntelligence, maturity, sociality, dominance, warmth, and credibility80
[19]1856Other reportCriminality89.5
[21]10,000Other reportBF89.11
[22]10,000Other reportBF90.94

Note. The table reflects the prediction accuracies of existing studies on personality prediction with images of the whole face. N photo is the number of samples, assessment is the personality assessment method of the study, M is the result for images of males, and F is the result for images of females.