Local Laplacian filters: edge-aware image processing with a Laplacian pyramid

By Sylvain Paris, Samuel W. Hasinoff, Jan Kautz. ACM Transactions on Graphics, 2011

627 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/f91afeda3f233c8f48fdb84120aecf4b1e45ad55

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Automated Crack Detection on Concrete Bridges (2016, methodology): “Our approach to multiscale analysis is to find the maximum response of over scale space using a Laplacian pyramid [30]–[32].”
Learning Recursive Filters for Low-Level Vision via a Hybrid Neural Network (2016, background): “In addition, the proposed model is much smaller and faster due to its hybrid structure, which can be used to accelerate more conventional algorithms, e.g., region covariance filter (RegCov) [39] and local laplacian filter (LLF) [40].”
Deep bilateral learning for real-time image enhancement (2017, methodology): “. the Local Laplacian filter [Paris et al. 2011] – an edge-preserving, multi-scale (yet non-scale-invariant) operator used for detail enhancement (we use two different strengths for the effect), ACM Transactions on Graphics, Vol. 36, No. 4, Article 118.”
Fast Image Processing with Fully-Convolutional Networks (2017, methodology): “Techniques developed in the last decade can dramatically enhance detail [24, 69, 26, 28, 60], transform the image by applying a master photographer’s style [7, 5], smooth the image for the purpose of abstraction [73, 76, 79], and eliminate the effects of atmospheric scattering [25, 35, 27, 9].”
Learned perceptual image enhancement (2017, methodology): “This loss is built on a no-reference quality pre-5 (a) Input (b) Reference [28] (c) CAN( L 2 ) [5] (d) CAN( L 2 +NIMA) Figure 7: Comparison of local detail enhancement operators in dark regions.”
Fast and Accurate Image Super-Resolution with Deep Laplacian Pyramid Networks (2017, positive): “vision tasks, including image blending [30], texture synthesis [31], edge-aware filtering [32] and semantic segmentation [33].”
Guided Image Filtering (2010, background): “Recently, Paris et al. [48] proposed manipulating the coefficients of the Laplacian Pyramid around each pixel for edge-aware filtering.”
Coherent Online Video Style Transfer (2017, background): “[34] extend the Gaussian kernel in bilateral filtering and mean-shift clustering to the temporal domain for video applications.”
LR3M: Robust Low-Light Enhancement via Low-Rank Regularized Retinex Model (2020, methodology): “Later on, edge-aware tone mapping methods such as edge-preserving decomposition [21] and local Laplacian filtering [22] are proposed to alleviate block artifacts.”
Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution (2017, methodology): “The Laplacian pyramid has been used in a wide range of applications, such as image blending [4], texture synthesis [14], edge-aware filtering [24] and semantic segmentation [11, 25].”
Single image portrait relighting (2019, methodology): “Local tone-mapping, which aims to decrease the dynamic range of an input image while preserving its local contrast, can also be thought of as a coarse approximation to relighting, and has been explored through classic algorithms [Farbman et al. 2008; Paris et al. 2011] and learning-based approaches ”