关键词：计算机视觉问题；三维结构；二维结构
摘 要：Recovering these same properties from a singleimage seems almost impossible in comparison——there are an infinite number of shapes, paint, and lights that exactly reproduce a single image. However, certain explanations are morelikely than others: surfaces tend to be smooth, paint tends to be uniform, and illumination tends to be natural. We therefore pose this problem as one of statistical inference, and define an optimization problem that searches for the most likely explanation of a single image.Ourmodel, which we call "SIRFS", can be viewed as a superset of several classic computer vision problems (shape-from-shading, intrinsic images, color constancy, illumination estimation, etc) and outperforms all previous solutions to those constituent problems. Recovering these same properties from a singleimage seems almost impossible in comparison——there are an infinite number of shapes, paint, and lights that exactly reproduce a single image. However, certain explanations are morelikely than others: surfaces tend to be smooth, paint tends to be uniform, and illumination tends to be natural. We therefore pose this problem as one of statistical inference, and define an optimization problem that searches for the most likely explanation of a single image.Ourmodel, which we call "SIRFS", can be viewed as a superset of several classic computer vision problems (shape-from-shading, intrinsic images, color constancy, illumination estimation, etc) and outperforms all previous solutions to those constituent problems.