Multispectral Image Intrinsic Decomposition via Low Rank Constraint
This addresses intrinsic decomposition for multispectral images, which is useful for computer vision tasks like recolorization and segmentation, but it is incremental as it adapts an existing RGB method to a new domain.
The paper tackles the problem of decomposing shading and reflectance from a single multispectral image by proposing a Low Rank Multispectral Image Intrinsic Decomposition model (LRIID), which extends the Retinex model and uses a low rank constraint to reduce ill-posedness, with experiments on a dataset of 12 images demonstrating effectiveness.
Multispectral images contain many clues of surface characteristics of the objects, thus can be widely used in many computer vision tasks, e.g., recolorization and segmentation. However, due to the complex illumination and the geometry structure of natural scenes, the spectra curves of a same surface can look very different. In this paper, a Low Rank Multispectral Image Intrinsic Decomposition model (LRIID) is presented to decompose the shading and reflectance from a single multispectral image. We extend the Retinex model, which is proposed for RGB image intrinsic decomposition, for multispectral domain. Based on this, a low rank constraint is proposed to reduce the ill-posedness of the problem and make the algorithm solvable. A dataset of 12 images is given with the ground truth of shadings and reflectance, so that the objective evaluations can be conducted. The experiments demonstrate the effectiveness of proposed method.