Crossing You in Style

Lee, Cheng-Che; Lin, Wan-Yi; Shih, Yen-Ting; Kuo, Pei-Yi; Su, Li

doi:10.1145/3394171.3413624

Cited by 8 publications

(6 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, we emphasize that the implemented pipeline produces paintings that also show expressive content and shapes. We provide some details in Figure 7; such images show high potential for the artistic possibilities and applications of this pipeline, which qualitatively outperforms the results obtained in [16] and in the other related works shown in Figure 2. Compared to [30,33,2], our paintings stand out for two main reasons: their content is generated ad-hoc on inputted EEG signals, and we have not pre-defined stylistic norms to express emotions.…”

Section: Example Experimentsmentioning

confidence: 60%

“…StyleGAN2 [11] is considered among the state-of-theart models to generate high-quality photorealistic images, and it is our reference architecture for this work. In the field of GAN conditioned with external data, the authors of [16] designed a model that generates paintings related to the composition epoch of an inputted piece of music. The music pieces and the paintings are related based on their historical epoch, which could be a simplistic and unrealistic correlation criterion.…”

Section: Background and Related Workmentioning

confidence: 99%

“…2: Example pictures from some of the related works. Top right [30], medium right [33], left [2], bottom [16]. Images are taken from the relative papers, and from the website in the case of [30].…”

Section: Preparation Of Datasetsmentioning

confidence: 99%

“…To shorten the time required by the training process, we provide the possibility of utilizing two extra losses (introduced in [16]): L cl and L sty , visible in Figure 4. L cl is obtained through an auxiliary classifier that identifies the emotion conveyed by the produced paintings and checks whether it is equal to the same emotion as the label of the inputted EEG wave.…”

Section: Extra Lossesmentioning

confidence: 99%

See 3 more Smart Citations

Translating Emotions from EEG to Visual Arts

Riccio

Galati

Zuluaga

et al. 2022

Artificial Intelligence in Music, Sound, Art and Design

View full text Add to dashboard Cite

Exploring the potentialities of artificial intelligence (AI) in the world of arts is fundamental to understand and define how this technology is shaping our creativity. We propose a system that generates emotionally expressive paintings from EEG signals. The emotional information, encoded from the signals through a graph neural network, is inputted to a generative adversarial network (GAN), trained on a dataset of paintings. The design and experimental choices at the base of this work rely on the understanding that emotions are hard to define and formalize. Despite this, the proposed results witness an interaction between an AI system and a human, capable of producing an original and artistic re-interpretation of emotions. These results have a promising potential for AI technologies applied to visual arts.

show abstract

Section: Example Experimentsmentioning

confidence: 60%

Section: Background and Related Workmentioning

confidence: 99%

Section: Preparation Of Datasetsmentioning

confidence: 99%

Section: Extra Lossesmentioning

confidence: 99%

See 2 more Smart Citations

Translating Emotions from EEG to Visual Arts

Riccio

Galati

Zuluaga

et al. 2022

Artificial Intelligence in Music, Sound, Art and Design

View full text Add to dashboard Cite

show abstract

“…15 further improved the quality of style transfer and achieved very good results. Recent style transfer methods have achieved good results and have been successfully applied in many fields, such as decorated logo generation, 16 music, 17 and font creation, 18 but they have not been applied to digital printing. Therefore, the application of style transfer methods in digital printing field to solve the problem of high design cost in personalized production still offers many research and development possibilities.…”

mentioning

confidence: 99%

Digital printing image generation method based on style transfer

Su,

Zhao,

Zhang

et al. 2023

Textile Research Journal

View full text Add to dashboard Cite

Digital printing has been widely used in textile printing production. In the process of designing digital printing patterns, an image generative model is needed to assist in obtaining more diversified patterns. However, the current model involves large storage space and high computing cost, which affects the promotion of digital printing customized production. To solve the problem, this article proposes a digital printing image generation method based on style transfer. Firstly, a style transfer method based on exact feature distribution matching is constructed to realize the accurate matching from image content to style features. And a balanced loss function is used to enhance the universality of the proposed method. Furthermore, knowledge distillation is introduced to compress the method proposed to reduce the hardware requirements when processing high-resolution digital printing images. Finally, a segmented training strategy is proposed to solve the performance degradation caused by model compression. The experimental results show that when processing images with a resolution of 3000 × 3000, the storage capacity of the model is only 2.68 MB and only 0.20 TFLOPs is required. The maximum processing resolution is more than 8K. The pattern obtained by this model is of high quality and can meet the needs of digital printing production.

show abstract

Robust sound-guided image manipulation

Lee,

Chi,

et al. 2024

Neural Networks

View full text Add to dashboard Cite

Crossing You in Style

Cited by 8 publications

References 19 publications

Translating Emotions from EEG to Visual Arts

Translating Emotions from EEG to Visual Arts

Digital printing image generation method based on style transfer

Robust sound-guided image manipulation

Contact Info

Product

Resources

About