Reconstruction of Textured Humans From Partial Scans

The goal of the project was to make IF-Net and Texture IF-Net by Chibane et al to work with real-world data scanned by a Kinect camera. The biggest contribution of the project was to prepare a new data set to train on top of the pre-trained Single View Reconstruction model. This was done to prepare the model to reconstruct a human which would be scanned by a Kinect camera from any side views.