Robotic Chinese Calligraphy with Human Preference

Chao, Fei; Lyu, Jitu; Wu, Ruiqi; Gao, Xingen; Zhou, Changle; Yang, Longzhi; Lin, Chunshui; Shang, Changjing

doi:10.1109/smartworld-uic-atc-scalcom-iop-sci.2019.00105

Cited by 4 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Existing intelligent robot writing mainly includes stroke models based on GAN, RNN and LSTM [8], [9], [20]- [23], Auto-Encoder [24] and deep reinforcement learning [6], [7], [16]- [18].…”

Section: B Neural Generator For Robotic Chinese Calligraphymentioning

confidence: 99%

“…GANCC Robot [9] improved the traditional GAN framework, added the type of stroke and the latent code information representing the characteristics of different strokes. The RCCHP [20] framework proposed the random sample as input of G network to obtain the position of the stroke trajectory, and the output was the trajectory distribution of Chinese characters. The G network designed by Wu R et al [21] used U-Net structure.…”

Section: B Neural Generator For Robotic Chinese Calligraphymentioning

confidence: 99%

See 1 more Smart Citation

CCD-BSMG: Composite-Curve-Dilation-Based Brush Stroke Model Generator for Robotic Chinese Calligraphy

Guo,

Ye,

Yan

2023

IEEE Access

View full text Add to dashboard Cite

Brush stroke training models for Chinese brush play an important role in robotic Chinese calligraphy as the basis of stroke generation. How to combine the end-to-end method and physical model needs to be further clarified. As a large number of brush strokes collected by robotic arm for learning and training lacks operability. The simulated brush stroke model can be used as a source of training datasets for deep learning and training instead of real samples collected by robot. With different combinations of physical parameters of the robotic arm, a sufficient amount of sample data can be generated by the simulated brush stroke model and then to train a generator based on the datasets. In this way, we propose a composite-curvedilation brush stroke model generator (CCD-BSMG) based on our composite-curve-dilation brush stroke model (CCD-BSM), which was formed by composite curve and morphological dilation directly according to the physical parameters of robotic arm and writing posture of the brush without a large number of samples for parameter estimation. The CCD-BSMG can generate the graphics with the dataset simulated by CCD-BSM for deep learning and training. Furthermore, with the output parameters of CCD-BSMG, the images reconstructed by robotic arm can have a better performance. Compared with other stroke generative models based on disordered pixels, our generator is based on parameterized brush strokes and provides a better foundation for robotic writing in deep learning or other fields. Compared with existing model and real stroke graphics written by robots, the results of several experiments prove that the proposed CCD-BSMG can generate stroke graphics well and show that it outperformed state-of-the-art stroke models. The results demonstrate the advantages of our proposed model in terms of high similarity and especially show the robustness and efficacy.INDEX TERMS stroke training model, Chinese brush, robotic Chinese calligraphy, brush stroke neural generator

show abstract

“…Existing intelligent robot writing mainly includes stroke models based on GAN, RNN and LSTM [8], [9], [20]- [23], Auto-Encoder [24] and deep reinforcement learning [6], [7], [16]- [18].…”

Section: B Neural Generator For Robotic Chinese Calligraphymentioning

confidence: 99%

Section: B Neural Generator For Robotic Chinese Calligraphymentioning

confidence: 99%

CCD-BSMG: Composite-Curve-Dilation-Based Brush Stroke Model Generator for Robotic Chinese Calligraphy

Guo,

Ye,

Yan

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Recently, DL has pushed such landscape‐making further towards commercialization and customization, thus benefiting the sharing of cocreation experiences among Human/AI in both real and virtual spaces. For instance, a calligraphy robot learns to write a customized style of calligraphy according to the user's esthetic preference (Chao et al., 2019). In this experiment, the calligraphy robot was trained by GAN to gain the ability to write different strokes.…”

Section: Revealing Socio‐spatial Practice Underlying Ai‐made CCLmentioning

confidence: 99%

When artificial intelligence comes to the Chinese calligraphic landscape: The coming transformation

Jiang

Jin

et al. 2022

Geography Compass

View full text Add to dashboard Cite

The rise of Artificial intelligence (AI) heralds potentially profound impact on the Chinese calligraphic landscape (CCL). Considering AI's increasing agential capacities, the anthropocentric conception of CCL that presupposes the priority of human identities, emotion, and creative work has been challenged. However, geographers remain quiet about AI-induced transformations up to date. To fill the research gap, this paper seeks to infuse more-than-human geographies into CCL. By taking a post-human approach, cultural geographers would have a novel understanding of human being in the creation of CCL. This paper initially discusses three prominent changes brought by deep learning (DL) in such landscape: a new ontological actor, transitory, and represented space. Responding to these transformations, the paper reconceptualizes the CCL as a post human term and unravels socio-spatial practices and diverse more-thanhuman geographies beneath such landscapes through three recent foci, namely robotic approaches to the CCL via DL, modeling experience brought about by AI, and human-AI collaboration for the creation of the CCL. Ultimately, this paper inspires geographers to profoundly comprehend CCLs in an era of AI. Through all these attempts, this paper advances insights into CCLs as more-than-humans-made.

show abstract

“…However, correcting errors in behaviors during the learning process requires more time to perform multiple tasks accurately. In recent years, the learning system of the Chinese character writing robot has the following approaches: generating the stroke trajectory by using a generative adversarial network (GAN) [1]- [3] or by using a control polygon [4], [5], converting the stroke trajectory from the image [6], modeling the stroke trajectory according to the font image [7]- [10], and using the perception system to obtain the course of the stroke [11]- [14], and so on.…”

Section: Introductionmentioning

confidence: 99%

A Chinese Calligraphy-Writing Robotic System Based on Image-to-Action Translations and a Hypothesis Generation Net

Hsu

Chien

et al. 2023

IEEE Access

View full text Add to dashboard Cite

This paper attempts to use a delta robot's structure and reliable coordinates to develop a selflearning Chinese calligraphy-writing system that requires precise control. Ideally, to achieve human-like behavior, a delta robot can learn stroke trajectories autonomously and present the stroke beauty of calligraphy characters. Unfortunately, state-of-the-art approaches have not yet considered the presentation of stroke beauty resulting from angles of rotation and tilt of the brush. This paper presents an integrated system consisting of a stroke processing module, a hypothesis generation net (HGN) learning model, a delta robot, and an image capture module. Our approach utilizes both the stroke trajectories from the stroke processing module and angles information from the HGN learning model to automatically produce five degrees of freedom action instructions. Based on the instructions, the delta robot completes calligraphy writing. Then, the image capture module provides feedback to the writing system for error calculation and coordinate correction. We utilize the mean absolute percentage error to verify the performance of the writing results. A correction algorithm and linear regression were used to improve the error correction results (less than 2% error). After several cycles, the written results approached the target sample finally. Consequently, the written results produced by the delta robot prove that our proposed system is capable of self-learning and correction.

show abstract

Robotic Chinese Calligraphy with Human Preference

Cited by 4 publications

References 13 publications

CCD-BSMG: Composite-Curve-Dilation-Based Brush Stroke Model Generator for Robotic Chinese Calligraphy

CCD-BSMG: Composite-Curve-Dilation-Based Brush Stroke Model Generator for Robotic Chinese Calligraphy

When artificial intelligence comes to the Chinese calligraphic landscape: The coming transformation

A Chinese Calligraphy-Writing Robotic System Based on Image-to-Action Translations and a Hypothesis Generation Net

Contact Info

Product

Resources

About