We observe the following positive effects for face recognition and facial landmark detection tasks: 1) Priming with synthetic face images improves the performance consistently across all benchmarks because it reduces the negative effects of biases in the training data.
We create a probabilistic joint face-skull model and show how to obtain a distribution of plausible face shapes given a skull shape.
In our experiments with an off-the-shelf face recognition software we observe the following phenomena: 1) The amount of real training data needed to train competitive deep face recognition systems can be reduced significantly.
4) We uncover a main limitation of current DCNN architectures, which is the difficulty to generalize when different identities to not share the same pose variation.
Therefore, we propose to learn self-shadowing for Morphable Model parameters directly with a linear model.