FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4321512602> ?p ?o ?g. }

• W4321512602 abstract "We present a Body Measurement network (BMnet) for estimating 3D anthropomorphic measurements of the human body shape from silhouette images. Training of BMnet is performed on data from real human subjects, and augmented with a novel adversarial body simulator (ABS) that finds and synthesizes challenging body shapes. ABS is based on the skinned multiperson linear (SMPL) body model, and aims to maximize BMnet measurement prediction error with respect to latent SMPL shape parameters. ABS is fully differentiable with respect to these parameters, and trained end-to-end via backpropagation with BMnet in the loop. Experiments show that ABS effectively discovers adversarial examples, such as bodies with extreme body mass indices (BMI), consistent with the rarity of extreme-BMI bodies in BMnet's training set. Thus ABS is able to reveal gaps in training data and potential failures in predicting under-represented body shapes. Results show that training BMnet with ABS improves measurement prediction accuracy on real bodies by up to 10%, when compared to no augmentation or random body shape sampling. Furthermore, our method significantly outperforms SOTA measurement estimation methods by as much as 3x. Finally, we release BodyM, the first challenging, large-scale dataset of photo silhouettes and body measurements of real human subjects, to further promote research in this area. Project website: https://adversarialbodysim.github.io." @default.
• W4321512602 created "2023-02-24" @default.
• W4321512602 creator A5003241140 @default.
• W4321512602 creator A5025958423 @default.
• W4321512602 creator A5027545499 @default.
• W4321512602 creator A5050679938 @default.
• W4321512602 creator A5064602017 @default.
• W4321512602 creator A5070528616 @default.
• W4321512602 creator A5075793611 @default.
• W4321512602 date "2022-09-01" @default.
• W4321512602 modified "2023-09-23" @default.
• W4321512602 title "Human Body Measurement Estimation with Adversarial Augmentation" @default.
• W4321512602 cites W1967554269 @default.
• W4321512602 cites W1988539193 @default.
• W4321512602 cites W1989191365 @default.
• W4321512602 cites W2134484928 @default.
• W4321512602 cites W2150297019 @default.
• W4321512602 cites W2431874326 @default.
• W4321512602 cites W2470632190 @default.
• W4321512602 cites W2487365028 @default.
• W4321512602 cites W2561715562 @default.
• W4321512602 cites W2562703451 @default.
• W4321512602 cites W2603766943 @default.
• W4321512602 cites W2749324691 @default.
• W4321512602 cites W2798637590 @default.
• W4321512602 cites W2916801517 @default.
• W4321512602 cites W2954884160 @default.
• W4321512602 cites W2962754033 @default.
• W4321512602 cites W2962997291 @default.
• W4321512602 cites W2963068442 @default.
• W4321512602 cites W2963515833 @default.
• W4321512602 cites W2963857521 @default.
• W4321512602 cites W2963873475 @default.
• W4321512602 cites W2963907666 @default.
• W4321512602 cites W2963918968 @default.
• W4321512602 cites W2963995996 @default.
• W4321512602 cites W2965523038 @default.
• W4321512602 cites W2968993450 @default.
• W4321512602 cites W2972504598 @default.
• W4321512602 cites W2978956737 @default.
• W4321512602 cites W2981637078 @default.
• W4321512602 cites W2982667067 @default.
• W4321512602 cites W2988856441 @default.
• W4321512602 cites W2990173985 @default.
• W4321512602 cites W2990747716 @default.
• W4321512602 cites W2997691263 @default.
• W4321512602 cites W2998254781 @default.
• W4321512602 cites W3009290327 @default.
• W4321512602 cites W3035291735 @default.
• W4321512602 cites W3035581100 @default.
• W4321512602 cites W3100388886 @default.
• W4321512602 cites W3107087058 @default.
• W4321512602 cites W3107471220 @default.
• W4321512602 cites W3107942585 @default.
• W4321512602 cites W3108691183 @default.
• W4321512602 cites W3109877674 @default.
• W4321512602 cites W3110631859 @default.
• W4321512602 cites W3154906787 @default.
• W4321512602 cites W3164632987 @default.
• W4321512602 cites W3165265377 @default.
• W4321512602 cites W3166523793 @default.
• W4321512602 cites W3175199633 @default.
• W4321512602 cites W3176070262 @default.
• W4321512602 cites W3177864674 @default.
• W4321512602 cites W3203745449 @default.
• W4321512602 cites W3204956438 @default.
• W4321512602 cites W3205884165 @default.
• W4321512602 cites W3215052027 @default.
• W4321512602 cites W4214517305 @default.
• W4321512602 cites W4214586188 @default.
• W4321512602 cites W4214684804 @default.
• W4321512602 cites W4214770715 @default.
• W4321512602 cites W4221142859 @default.
• W4321512602 cites W4225964194 @default.
• W4321512602 cites W4231018136 @default.
• W4321512602 cites W4312261745 @default.
• W4321512602 cites W4312767894 @default.
• W4321512602 cites W4312809617 @default.
• W4321512602 doi "https://doi.org/10.1109/3dv57658.2022.00034" @default.
• W4321512602 hasPublicationYear "2022" @default.
• W4321512602 type Work @default.
• W4321512602 citedByCount "0" @default.
• W4321512602 crossrefType "proceedings-article" @default.
• W4321512602 hasAuthorship W4321512602A5003241140 @default.
• W4321512602 hasAuthorship W4321512602A5025958423 @default.
• W4321512602 hasAuthorship W4321512602A5027545499 @default.
• W4321512602 hasAuthorship W4321512602A5050679938 @default.
• W4321512602 hasAuthorship W4321512602A5064602017 @default.
• W4321512602 hasAuthorship W4321512602A5070528616 @default.
• W4321512602 hasAuthorship W4321512602A5075793611 @default.
• W4321512602 hasBestOaLocation W43215126022 @default.
• W4321512602 hasConcept C106131492 @default.
• W4321512602 hasConcept C119857082 @default.
• W4321512602 hasConcept C121332964 @default.
• W4321512602 hasConcept C140779682 @default.
• W4321512602 hasConcept C154945302 @default.
• W4321512602 hasConcept C177264268 @default.
• W4321512602 hasConcept C199360897 @default.
• W4321512602 hasConcept C2778755073 @default.
• W4321512602 hasConcept C31972630 @default.

• next