Learning Depth Completion of Transparent Objects using Augmented Unpaired Data
Floris Erich,Bruno Leme,Noriaki Ando,Ryo Hanai,Yukiyasu Domae,Floris Erich,Bruno Leme,Noriaki Ando,Ryo Hanai,Yukiyasu Domae
We propose a technique for depth completion of transparent objects using augmented data captured directly from real environments with complicated geometry. Using cyclic adversarial learning we train translators to convert between painted versions of the objects and their real transparent counterpart. The translators are trained on unpaired data, hence datasets can be created rapidly and without an...
