Unveiling face images of a subject given his/her high-level representations\nextracted from a blackbox Face Recognition engine is extremely challenging. It\nis because the limitations of accessible information from that engine including\nits structure and uninterpretable extracted features. This paper presents a\nnovel generative structure with Bijective Metric Learning, namely Bijective\nGenerative Adversarial Networks in a Distillation framework (DiBiGAN), for\nsynthesizing faces of an identity given that person's features. In order to\neffectively address this problem, this work firstly introduces a bijective\nmetric so that the distance measurement and metric learning process can be\ndirectly adopted in image domain for an image reconstruction task. Secondly, a\ndistillation process is introduced to maximize the information exploited from\nthe blackbox face recognition engine. Then a Feature-Conditional Generator\nStructure with Exponential Weighting Strategy is presented for a more robust\ngenerator that can synthesize realistic faces with ID preservation. Results on\nseveral benchmarking datasets including CelebA, LFW, AgeDB, CFP-FP against\nmatching engines have demonstrated the effectiveness of DiBiGAN on both image\nrealism and ID preservation properties.\n