Mixing two synthetic face datasets with different generation pipelines improves face recognition on three of five benchmarks, but gains are small and confounded by unspecified subset selection.
Disentangled and controllable face image genera- tion via 3d imitative-contrastive learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing Domain Diversity in Synthetic Data Face Recognition with Dataset Fusion
Mixing two synthetic face datasets with different generation pipelines improves face recognition on three of five benchmarks, but gains are small and confounded by unspecified subset selection.