Editing Non-photorealistic rendering (section)

==2D==
[[File:Cradle Mountain Behind Dove Lake painted.jpg|thumb|300px|right|A non-photorealistic rendering of an existing 2D (photographic) image]]
[[File:Cradle Mountain Behind Dove Lake.jpg|thumb|300px|Original here]]
The input to a two dimensional NPR system is typically an image or video. The output is a typically an artistic rendering of that input imagery (for example in a watercolor, painterly or sketched style) although some 2D NPR serves non-artistic purposes e.g. data visualization.

The artistic rendering of images and video (often referred to as ''image stylization''<ref>{{cite book |doi=10.1007/978-1-4471-4519-6 |title=Image and Video-Based Artistic Stylisation |series=Computational Imaging and Vision |year=2013 |volume=42 |isbn=978-1-4471-4518-9 |s2cid=40656135 |editor1-first=Paul |editor1-last=Rosin |editor2-first=John |editor2-last=Collomosse }}{{pn|date=July 2022}}</ref>) traditionally focused upon heuristic algorithms that seek to simulate the placement of brush strokes on a digital canvas.<ref>{{cite journal |last1=Kyprianidis |first1=Jan Eric |last2=Collomosse |first2=John |last3=Wang |first3=Tinghuai |last4=Isenberg |first4=Tobias |title=State of the 'Art': A Taxonomy of Artistic Stylization Techniques for Images and Video |journal=IEEE Transactions on Visualization and Computer Graphics |date=May 2013 |volume=19 |issue=5 |pages=866–885 |doi=10.1109/TVCG.2012.160 |pmid=22802120 |s2cid=2656810 |url=https://hal.inria.fr/hal-00781502/file/Kyprianidis_2013_SAT.pdf }}</ref>

Arguably, the earliest example of 2D NPR is [[Paul Haeberli]]'s '[[paint by number|Paint by Numbers]]' at [[SIGGRAPH]] 1990. This (and similar interactive techniques) provide the user with a canvas that they can "paint" on using the cursor — as the user paints, a stylized version of the image is revealed on the canvas. This is especially useful for people who want to simulate different sizes of brush strokes according to different areas of the image.

Subsequently, basic image processing operations using gradient operators<ref>{{cite book |doi=10.1145/258734.258893 |chapter=Processing images and video for an impressionist effect |title=Proceedings of the 24th annual conference on Computer graphics and interactive techniques - SIGGRAPH '97 |year=1997 |last1=Litwinowicz |first1=Peter |pages=407–414 |isbn=978-0-89791-896-1 |s2cid=13139308 }}</ref> or statistical moments<ref>{{cite book |doi=10.1145/340916.340923 |chapter=An algorithm for automatic painterly rendering based on local source image approximation |title=Proceedings of the first international symposium on Non-photorealistic animation and rendering - NPAR '00 |year=2000 |last1=Shiraishi |first1=Michio |last2=Yamaguchi |first2=Yasushi |pages=53–58 |isbn=978-1-58113-277-9 |s2cid=16915734 }}</ref> were used to automate this process and minimize user interaction in the late nineties (although artistic control remains with the user via setting parameters of the algorithms). This automation enabled practical application of 2D NPR to video, for the first time in the living paintings of the movie ''[[What Dreams May Come (film)|What Dreams May Come]]'' (1998).

More sophisticated image abstractions techniques were developed in the early 2000s harnessing computer vision operators e.g. image salience,<ref>{{cite book |doi=10.1109/EGUK.2002.1011281 |citeseerx=10.1.1.7.5383 |chapter=Painterly rendering using image salience |title=Proceedings 20th Eurographics UK Conference |year=2002 |last1=Collomosse |first1=J.P. |last2=Hall |first2=P.M. |pages=122–128 |isbn=978-0-7695-1518-2 |s2cid=9610948 }}</ref> or segmentation<ref>{{cite book |doi=10.1145/508530.508545 |citeseerx=10.1.1.10.1761 |chapter=Artistic Vision |title=Proceedings of the second international symposium on Non-photorealistic animation and rendering - NPAR '02 |year=2002 |last1=Gooch |first1=Bruce |last2=Coombe |first2=Greg |last3=Shirley |first3=Peter |page=83 |isbn=978-1-58113-494-0 |s2cid=1146198 }}</ref> operators to drive stroke placement. Around this time, machine learning began to influence image stylization algorithms notably [[image analogy]]<ref>{{cite book |doi=10.1145/383259.383295 |citeseerx=10.1.1.119.5127 |chapter=Image analogies |title=Proceedings of the 28th annual conference on Computer graphics and interactive techniques - SIGGRAPH '01 |year=2001 |last1=Hertzmann |first1=Aaron |last2=Jacobs |first2=Charles E. |last3=Oliver |first3=Nuria |last4=Curless |first4=Brian |last5=Salesin |first5=David H. |pages=327–340 |isbn=978-1-58113-374-5 |s2cid=2201072 }}</ref> that could learn to mimic the style of an existing artwork.

The advent of [[deep learning]] has re-kindled activity in image stylization, notably with [[neural style transfer]] (NST) algorithms that can mimic a wide gamut of artistic styles from single visual examples. These algorithms underpin mobile apps capable of the same e.g. [[Prisma (app)|Prisma]]

In addition to the above stylization methods, a related class of techniques in 2D NPR address the simulation of artistic media. These methods include simulating the [[diffusion]] of ink through different kinds of [[paper]], and also of [[pigment]]s through water for simulation of [[watercolor]].