RPG-Palm: Realistic Pseudo-data Generation for Palmprint Recognition
This addresses a data scarcity problem for researchers and developers in biometrics, offering an incremental improvement in synthetic data generation for palmprint recognition.
The paper tackles the lack of large-scale public palmprint datasets by proposing a realistic pseudo-palmprint generation model, which improves state-of-the-art recognition performance by over 5% and 14% in specific metrics and outperforms methods using full real data with only 10% of it.
Palmprint recently shows great potential in recognition applications as it is a privacy-friendly and stable biometric. However, the lack of large-scale public palmprint datasets limits further research and development of palmprint recognition. In this paper, we propose a novel realistic pseudo-palmprint generation (RPG) model to synthesize palmprints with massive identities. We first introduce a conditional modulation generator to improve the intra-class diversity. Then an identity-aware loss is proposed to ensure identity consistency against unpaired training. We further improve the Bézier palm creases generation strategy to guarantee identity independence. Extensive experimental results demonstrate that synthetic pretraining significantly boosts the recognition model performance. For example, our model improves the state-of-the-art BézierPalm by more than $5\%$ and $14\%$ in terms of TAR@FAR=1e-6 under the $1:1$ and $1:3$ Open-set protocol. When accessing only $10\%$ of the real training data, our method still outperforms ArcFace with $100\%$ real training data, indicating that we are closer to real-data-free palmprint recognition.