Adaptive fast local Laplacian filters and its edge-aware application

We present a new approach for edge-aware image processing, inspired by the principle of local Laplacian filters and fast local Laplacian filters. In contrast to the previous methods that primarily rely on fixed intensity threshold, our method adopts an adaptive parameter selection strategy in different regions of the processing image. This adaptive parameter selection strategy allows different intensity thresholds and different amplitude magnification factors in different pixels, moreover, a different remapping functions are adopted to process each pixel. At the same time, we propose an efficient and flexible method for obtaining the representation of image local variation, and based on the representation to select local Laplacian filters parameters adaptively. Our experiments shows that high-quality results in the detail enhancement and detail smoothing can be produced by our methods.