Poisson image editing

By Patrick Pérez, Michel Gangnet, Andrew Blake. 2003

3,188 citations recorded by the source corpus.

Subfields: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

CNN-Generated Images Are Surprisingly Easy to Spot… for Now (2019, methodology): “This model uses an autoencoder to generate faces, and images undergo extensive post-processing steps, including Poisson image blending [35] with real content.”
Semantic Image Inpainting with Deep Generative Models (2016, methodology): “The minimization problem contains a quadratic term, which has a unique solution [31].”
Synthetic Data for Text Localisation in Natural Images (2016, methodology): “Finally, a text sample is rendered using a randomly selected font and transformed according to the local surface orientation; the text is blended into the scene using Poisson image editing [35].”
Image Inpainting for Irregular Holes Using Partial Convolutions (2018, methodology): “[10] uses fast marching [30] and Poisson image blending [21], while Yu et al.”
Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detection (2017, background): “Each of these modes add different image variations, e.g ., Poisson blending [37] smooths edges and adds lighting variations.”
SIFT Flow: Dense Correspondence across Scenes and Its Applications (2011, background): “SIFT flow attempts to match the query image by reshuffling the pixels in the candidate image.”
Modeling Visual Context is Key to Augmenting Object Detection Datasets (2018, methodology): “to prevent the detector to \detect artefacts" instead of detecting objects of interest. This is achieved by using various blending strategies to smooth object boundaries such as Poisson blending [34], and by adding \distractors" that are objects that do not belong to any of the dataset cat”
Editing in Style: Uncovering the Local Semantics of GANs (2020, methodology): “Figure 4: Unlike previous blending methods[24, 26], our method does not require images to be aligned or of similar scale.”
GP-GAN: Towards Realistic High-Resolution Image Blending (2017, methodology): “be implemented by using the divergence operator. Formally, we need to optimize the objective function defined as follows: H(xh)= P(xh)+ βC(xh). (5) P(xh)is inspired by the well-known Poisson Equation [23] and is defined as follows: P(xh)= ∫ T ∥divv−∆xh∥2 2 dt, (6) C(xh)is defined as follows: C(xh)= ”
Learning a No-Reference Quality Metric for Single-Image Super-Resolution (2016, methodology): “For each patch, we select the one from all results with highest quality scores and stitch all the selected patches together using the graph cut and Poisson blending [46] method (See Figure 16(d)).”
Using Color Compatibility for Assessing Image Realism (2007, background): “[14] propose to copy the object gradients and reintegrate to get the colors.”
Implicit Neural Representations with Periodic Activation Functions (2020, background): “Images can be seamlessly fused in the gradient domain [49].”
Fast Global Image Smoothing Based on Weighted Least Squares (2014, neutral): “The second class of existing EPS methods are based on global optimization formulations [13]–[17].”
As-Projective-As-Possible Image Stitching with Moving DLT (2013, background): “In the second group, seam cutting [1, 3] and Poisson blending [13] are influential.”
Non-uniform Deblurring for Shaken Images (2010, methodology): “The variational output (d) is estimated using the marginalization algorithm for the non-uniform case (calculated as 〈f〉q(f) then converted from gradients to intensities using Poisson reconstruction (Pérez et al. 2003)).”