Generation of a 3D model of the inside volume of shoes for e-commerce applications using industrial x-ray computed tomography

Wittmann, J.; Herl, Gabriel; Hiller, Jochen

doi:10.1088/2631-8695/ac43c8

Cited by 3 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FastkNN [152] FCNN [56] FNN [32] Fuzzylogic [106,113] GAN [71,75,99,129,134,141,155,251] GA-RF [65] GaussianMixtureModels [109,118] gradientboostingdecisiontree [42] GraphCNN [40,122] GraphDCNN [89] GRU [230] GSN [87] H-CNN [39,157] HOG [153] HypergraphNN [80] Imageprocessing [119,120,130,132,135,136,147] k-center [117] k-means [114,203,232,244] k-medoids [115] Kneser-Ney [34] kNN [54,137,222,235,237,250] LAC [107,116] Lassoregression …”

Section: Methods Citationsmentioning

confidence: 99%

“…Summary of purposes with references. Sentimentanalysis [42,44,[169][170][171][172][173][174][175][176][177][178][179][180] Virtualfitting [32,[129][130][131][132][133][134][135][136][138][139][140][141][142][143][144][145][146][149][150][151][152] Table 7. Summary of databases with references.…”

Section: R-cnn [59]mentioning

confidence: 99%

“…Collectedfromreviewsofretailshops [42,44,160,165,169,170,174,[178][179][180]183] Collectedfromscanner [46,132] Collectedfromsocialmedia [107,116,172,175,222] CollectedfromtheInternet [81,95,102,104,117,128,153,155,167,168,176,177,192,203,225,…”

Section: Databases Citationsmentioning

confidence: 99%

See 2 more Smart Citations

Artificial Intelligence in Business-to-Customer Fashion Retail: A Literature Review

Goti,

Querejeta-Lomas,

Almeida

et al. 2023

Mathematics

View full text Add to dashboard Cite

Many industries, including healthcare, banking, the auto industry, education, and retail, have already undergone significant changes because of artificial intelligence (AI). Business-to-Customer (B2C) e-commerce has considerably increased the use of AI in recent years. The purpose of this research is to examine the significance and impact of AI in the realm of fashion e-commerce. To that end, a systematic review of the literature is carried out, in which data from the Web Of Science and Scopus databases were used to analyze 219 publications on the subject. The articles were first categorized using AI techniques. In the realm of fashion e-commerce, they were divided into two categories. These categorizations allowed for the identification of research gaps in the use of AI. These gaps offer potential and possibilities for further research.

show abstract

Section: Methods Citationsmentioning

confidence: 99%

Section: R-cnn [59]mentioning

confidence: 99%

See 1 more Smart Citation

Artificial Intelligence in Business-to-Customer Fashion Retail: A Literature Review

Goti,

Querejeta-Lomas,

Almeida

et al. 2023

Mathematics

View full text Add to dashboard Cite

show abstract

“…All three methods require manual unboxing, removal of packaging material, preparation of the shoe for measurement, and manual measurement. A recent study by Wittmann et al proposes to extract the shoe's internal volume (SIV) from computed tomography scans and then model the key features of the shoe [16]. Therefore, this approach uses boxed shoes and is theoretically suitable for large-scale automation.…”

Section: Related Workmentioning

confidence: 99%

Three Step Volumetric Segmentation for Automated Shoe Fitting

Leipert¹,

Herl²,

Stebani³

et al. 2023

eJNDT

View full text Add to dashboard Cite

This work presents a three-step segmentation process based on Convolutional Neural Networks. The task is to identify the different parts of shoes from Computed Tomography scans of boxed pairs of shoes. The first step of the three-step algorithm uses a scaled-down volume image to separate the shoe material from its surroundings. The second step segments the shoe's inside volume, i.e. the space enclosed by shoe material. The third and last step splits the segmented shoe material into individual components: shoe upper material, outer and insole. The complete process employs CNNs derived from three-dimensional UNets. Residual SE UNet, Dense UNet, and Bottleneck Residual UNet are evaluated for the three steps. The architectures are modified for large receptive fields. The networks are trained and tested for each step separately and conjointly on CT scans comprising various shoe types. The test results inspire hope for using the process for automated segmentation and extraction of meshes from large batches of CT scans. In particular, the first step using a Residual SE UNet achieves an F1-score of 88.2 % for shoes and 58.9 % for the packing material. The second step segments the inside volume with an F1-score of 81.0 %. The third step segments the shoe into its components and achieves an F1-score for insole of 79.5 %, outer sole of 88.7 % and upper material of 81.3 %.

show abstract

“…Better fitting algorithms and exact recommendations could reduce return rates significantly. Wittmann et al developed an algorithm for this use case to digitize the shoe's inner volume and extract its surface mesh from boxed half shoes based on region growing and active contours [9].…”

Section: Related Workmentioning

confidence: 99%

Automated Shoe Metrology by X-Ray Computed Tomography

LEIPERT,

HERL,

MUELLER

et al. 2023

Proceedings of 3DBODY.TECH 2023 - 14th International Conference and Exhibition on 3D Body Scanning and Processing Technologies,

View full text Add to dashboard Cite

Automation of shoe metrology is crucial to providing fit information for shoes on a large scale. Here, we examine a segmentation technique to extract the inner shoe volume (ISV) from Computed Tomography (CT) data-the proposed approach leverages artificial neural networks to extract shoe parts for automated metrology precisely. The neural network architecture is customized to facilitate the extraction of ISV by integrating spatial attention mechanisms. Furthermore, a neural network segmentation algorithm removes filler materials virtually. This process yields enhancements of 1.3% in F1-score through material removal and an additional 1.4% through the incorporation of spatial attention. Notably, spatial attention mechanisms yield improved outcomes at the aperture of the shoe. The elimination of filler materials reduces false positive segmentations. The segmentation outcomes are utilized to generate surface meshes. These are compared to surface meshes derived from annotated data. We measure an average Hausdorff distance between annotated and labelled data of 2.1 mm. The discrepancy is primarily attributed to deformations and artifacts. On both sets, we measure the effective shoe length. Precision and accuracy metrics for the extracted measurement from ANN-segmented data attain 0.8 mm and 1.8 mm, respectively. For meshes obtained from label data, the precision is 0.2 mm, and the accuracy is 2.5 mm. Our findings underscore the accuracy of the extracted shoe interior volumes, rendering them suitable for metrological applications. Limitations include unsolved issues with separation reliability and deformation.

show abstract

Generation of a 3D model of the inside volume of shoes for e-commerce applications using industrial x-ray computed tomography

Cited by 3 publications

References 17 publications

Artificial Intelligence in Business-to-Customer Fashion Retail: A Literature Review

Artificial Intelligence in Business-to-Customer Fashion Retail: A Literature Review

Three Step Volumetric Segmentation for Automated Shoe Fitting

Automated Shoe Metrology by X-Ray Computed Tomography

Contact Info

Product

Resources

About