Inverse Transport Networks
CoRR, Volume abs/1809.10820, 2018.
We introduce inverse transport networks as a learning architecture for inverse rendering problems where, given input image measurements, we seek to infer physical scene parameters such as shape, material, and illumination. During training, these networks are evaluated not only in terms of how close they can predict groundtruth parameter...More
Full Text (Upload PDF)
PPT (Upload PPT)