In his thesis, Marc has introduced numerous highly original, highly impactful, and far-reaching contributions to the topic of 3D and 4D human reconstruction and synthesis from video. The presented new methods to combine model-based and learning-based concepts in end-to-end-trainable combined architectures and the new methods to train them purely on 2D image data rather than explicit 3D data as supervision are groundbreaking and highly relevant beyond the core topic of virtual humans.
Since 2015, the Deutsche Arbeitsgemeinschaft für Mustererkennung e.V has granted the DAGM MVTec Dissertation Award. The Dissertation Award honors an outstanding dissertation successfully completed in the preceding year in the fields of pattern recognition, image processing, machine vision, and machine learning.