A data-driven robotic Chinese calligraphy system using convolutional auto-encoder and differential evolution

Gao, Xingen; Zhou, Changle; Chao, Fei; Yang, Longzhi; Lin, Chih Min; Xu, Tao; Shang, Changjing; Shen, Qiang

doi:10.1016/j.knosys.2019.06.010

Cited by 22 publications

(11 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, over the same feature vectors obtained by CAEs model (which is adopted in PAADA for data compression), we experimentally compare the proposed PAADA with existing adaptive sampling algorithms EDSAS [17] and DDASA [19] since they are the latest methods. The reason of choosing CAEs model for image data compression is that it performs better than existing widely adopted data compression algorithms PCA [43] and AEs model [44], [45] (as shown in Section VII-B).…”

Section: ) Algorithms For Comparisonmentioning

confidence: 99%

PAADA: Physical-World-Aware Approximate Data Acquisition for Image Sensor Networks

et al. 2020

IEEE Access

View full text Add to dashboard Cite

To observe the complicated physical world, sensor networks are widely used for data collection. Moreover, due to the limited energy, storage and computation capacity of sensor nodes, approximate data acquisition in adaptive sampling manner is a wide choice. Nevertheless, existing data acquisition methods are most designed for univariate data (e.g., temperature), and thus not applicable to image data with high dimensions and complex structures. In this paper, we propose a framework of Physical-world-Aware Adaptive Data Acquisition (PAADA) for image sensor networks, to sample data adaptively with prespecified error bound. First, based on the convolutional autoencoders (CAEs), PAADA compresses the highdimensional image data into a feature vector with a handful of hidden variables which compactly capture the key features of the image data. Second, PAADA designs a Physical-world-aware Adaptive Sampling (PAS) algorithm based on the Hermitee interpolation. Under the feature space, the PAS algorithm adjusts the sampling frequency automatically by considering the change trend of the feature vector. In addition, the feature vectors at non-sampling time points can be recovered with O () approximation guarantee to the ground truths. Next, PAADA recovers the image data at non-sampling time points based on the recovered feature vectors. Finally, for each sensor, PAADA returns an image series composed of sampled images (at sampling time points) and approximate images (at non-sampling time points). Experiments on realworld datasets demonstrate that the proposed PAADA has high performance in both accuracy and energy consumption. INDEX TERMS Image sensor networks, adaptive data acquisition, physical-world-aware approximation.

show abstract

Section: ) Algorithms For Comparisonmentioning

confidence: 99%

PAADA: Physical-World-Aware Approximate Data Acquisition for Image Sensor Networks

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Most of the existing robotic writing systems do not have a realistic aesthetic evaluation mechanism to critically assess the writing performance [3]. Instead, several studies adopted the Fréchet inception distances (FID) or the restoring accuracy of an Autoencoder network to represent the aesthetic performance [4], [26], [27]. These methods well examined the writing results based on distribution, but often ignored the structural information of letters or numerals representing diversity.…”

Section: Introductionmentioning

confidence: 99%

Solving Robotic Trajectory Sequential Writing Problem via Learning Character’s Structural and Sequential Information

Guo

Chao

et al. 2024

IEEE Trans. Cybern.

Self Cite

View full text Add to dashboard Cite

The writing sequence of numerals or letters often affects aesthetic aspects of the writing outcomes. As such, it remains a challenge for robotic calligraphy systems to perform, mimicking human writers' implicit intention. This paper presents a new robot calligraphy system that is able to learn writing sequences with limited sequential information, producing writing results compatible to human writers with good diversity. In particular, the system innovatively applies a Gated Recurrent Unit (GRU) network to generate robotic writing actions with the support of a pre-labeled trajectory sequence vector. Also, a new evaluation method is proposed that considers the shape, trajectory sequence, and structural information of the writing outcome, thereby helping ensure the writing quality. A swarm optimization algorithm is exploited to create an optimal set of parameters of the proposed system.The proposed approach is evaluated using Arabic numerals, and the experimental results demonstrate the competitive writing performance of the system against state-of-the-art approaches regarding multiple criteria (including FID, MAE, PSNR, SSIM, and PerLoss), as well as diversity performance concerning variance and entropy. Importantly, the proposed GRU-based robotic motion planning system, supported with swarm optimization can learn from a small dataset, while producing calligraphy writing with diverse and aesthetically pleasing outcomes.

show abstract

“…Chinese is a morphosyllabic writing system comprising more than 50,000 characters, of which about 5,000 are commonly used in everyday life ( Gao et al, 2019 ). Pinyin, a phonetic transcription of Chinese characters, was established by the Chinese government in 1958 ( Zhou, 1958 ) to support the pronunciation of characters.…”

Section: Introductionmentioning

confidence: 99%

Does Early Exposure to Chinese–English Biliteracy Enhance Cognitive Skills?

et al. 2022

View full text Add to dashboard Cite

Clarifying the effects of biliteracy on cognitive development is important to understanding the role of cognitive development in L2 learning. A substantial body of research has shed light on the cognitive factors contributing to biliteracy development. Yet, not much is known about the effect of the degree of exposure to biliteracy on cognitive functions. To fill this research void, we measured three categories of biliteracy skills (language, reading, and cognitive–linguistic skills in both Chinese and English) jointly and investigated the effects of biliteracy skill performance in these three categories on cognitive skills (working memory and attentional control) in Chinese L1 children who were exposed to English as L2 beginning at age 3 (“early”) or in grade 3 (“late”). In this cross-sectional study, 10 parallel Chinese and English language, reading, and cognitive–linguistic measures were administered to emerging Chinese–English bilingual children in grade 3 (n = 178) and grade 6 (n = 176). The results revealed that early exposure to Chinese–English biliteracy enhanced cognitive skills but with a cost of a slight delay in performance on Chinese L1 language skills in grade 3 (but not in grade 6). Importantly, the present findings suggest that, in addition to universal and developmental processes, the cumulative effect of early and sustained bi-scriptal exposure enhances working memory and attention in school children.

show abstract

A data-driven robotic Chinese calligraphy system using convolutional auto-encoder and differential evolution

Cited by 22 publications

References 37 publications

PAADA: Physical-World-Aware Approximate Data Acquisition for Image Sensor Networks

PAADA: Physical-World-Aware Approximate Data Acquisition for Image Sensor Networks

Solving Robotic Trajectory Sequential Writing Problem via Learning Character’s Structural and Sequential Information

Does Early Exposure to Chinese–English Biliteracy Enhance Cognitive Skills?

Contact Info

Product

Resources

About