Vinci: An Intelligent Graphic Design System for Generating Advertising Posters

Guo, Shunan; Jin, Zhuochen; Sun, Fuling; Li, Jingwen; Li, Zhaorui; Shi, Yang; Cao, Nan

doi:10.1145/3411764.3445117

Cited by 56 publications

(38 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Zheng et al [48] propose a deep generative model to synthesize layouts based on the semantics of visual elements. Vinci [12] uses a sequenceto-sequence generative model to match the visual elements and layouts for generating an aesthetic advertising poster. LayoutGAN [21] synthesizes layouts from different types of 2D visual elements by using a Generative Adversarial Network architecture.…”

Section: Graphic Design Layoutmentioning

confidence: 99%

“…There are two main reasons to use a Transformer-based model. First, as an information presentation could be formalized as a design sequence [12], it is reasonable to adopt a sequence model in our task. Compared with traditional sequence models like recurrent neural networks, the self-attention mechanism [35] in Transformer models can capture long-range contextual information in the whole layout design.…”

Section: Model Architecturementioning

confidence: 99%

See 1 more Smart Citation

Reverse-Engineering Information Presentations: Recovering Hierarchical Grouping from Layouts of Visual Elements

Shi¹,

Cui²,

Huang³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Visual elements in an information presentation are often spatially and semantically grouped hierarchically for effective message delivery. Studying the hierarchical grouping information can help researchers and designers better explore layout structures and understand design demographics. However, recovering hierarchical grouping is challenging due to a large number of possibilities for compositing visual elements into a single-page design. This paper introduces an automatic approach that takes the layout of visual elements as input and returns the hierarchical grouping as output. To understand information presentations, we first contribute a dataset of 23,072 information presentations with diverse layouts to the community. Next, we propose our technique with a Transformer-based model to predict relatedness between visual elements and a bottom-up algorithm to produce the hierarchical grouping. Finally, we evaluate our technique through a technical experiment and a user study with thirty designers. Results show that the proposed technique is promising.CCS Concepts: • Human-centered computing → Visualization design and evaluation methods.

show abstract

Section: Graphic Design Layoutmentioning

confidence: 99%

Section: Model Architecturementioning

confidence: 99%

Reverse-Engineering Information Presentations: Recovering Hierarchical Grouping from Layouts of Visual Elements

Shi¹,

Cui²,

Huang³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this paper we propose AI-assisted design, a general-purpose framework for cooperative assistants in design problems. Cooperative design assistance has been proposed for specific design problems [3], but no general framework currently exists. The framework is designed to support a wide range of interactions between assistant and designer, though our focus here will be on design change recommendations.…”

Section: Cooperative Design Assistancementioning

confidence: 99%

“…In computational drug design, there is an ongoing effort to speed up and automate the identification of promising drug molecules using computational methods [2], allowing scientists to find better cures faster. And in graphic design, generative deep learning meth-ods are helping designers create more appealing posters [3].…”

mentioning

confidence: 99%

Toward AI Assistants That Let Designers Design

Peuter¹,

Oulasvirta²,

Kaski³

2021

Preprint

View full text Add to dashboard Cite

AI for supporting designers needs to be rethought. It should aim to cooperate, not automate, by supporting and leveraging the creativity and problem-solving of designers. The challenge for such AI is how to infer designers' goals and then help them without being needlessly disruptive. We present AI-assisted design: a framework for creating such AI, built around generative user models which enable reasoning about designers' goals, reasoning, and capabilities.Computer

show abstract

“…Graphics spread information everywhere in the world, and it has become an expression language without national boundaries. On the one hand, modern print advertising is a new way to promote commercial advertising; on the other hand, it is also the expression of aesthetic symbolism in the market economy [ 14 ]. Graphic vision in modern print advertising design, as a combination of technology and art, embodies the diversity and richness of modern culture all the time.…”

Section: Introductionmentioning

confidence: 99%

Visual Health Analysis of Print Advertising Graphic Design Based on Image Segmentation and Few-Shot Learning

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Graphics innovation must adapt to the changing trends of the times in order to stay ahead of the curve. This article investigates the use of graphic vision in print advertising design, using image segmentation as a starting point and examines the current state of visual innovation in print advertising graphic design and its various expressions and applications. A series of homogeneous regions are generated and outlined using an image segmentation algorithm based on adaptive local threshold. The user then paints different colors on the area sets that make up the various targets. Finally, the image segmentation is completed by merging the color mark region sets. Automatic extraction of the initial curve of an active contour model, construction of an active contour model based on saliency and level set solution, automatic selection of training samples when a classifier is used for image segmentation, and so on are all problems that this method effectively solves. Experiments show that this algorithm not only satisfies users’ demands for more intuitive input and more accurate interactive image segmentation results but also enables multiregion and multitarget image segmentation with ease.

show abstract

Vinci: An Intelligent Graphic Design System for Generating Advertising Posters

Cited by 56 publications

References 31 publications

Reverse-Engineering Information Presentations: Recovering Hierarchical Grouping from Layouts of Visual Elements

Reverse-Engineering Information Presentations: Recovering Hierarchical Grouping from Layouts of Visual Elements

Toward AI Assistants That Let Designers Design

Visual Health Analysis of Print Advertising Graphic Design Based on Image Segmentation and Few-Shot Learning

Contact Info

Product

Resources

About