Intrinsic and invariant information are fundamental components of the image formation process. In fact, the image formation heavily depends on the geometry of the scene, object material, and illumination. To decompose the image into its constituent parts, well-awareness of the geometry information in the scene and the illumination is a necessity. Previous methods focussed on understanding the scence information from a single or series of images. However, the geometry part is still an unexplored area. In this project, we develop methods to better address the decomposition task by merging the implicit/explicit geometry information and RGB images.