High-Accuracy Automatic Person Segmentation with Novel Spatial Saliency Map

Xi, Weijuan; Chen, Jianhang; Lin, Qian; Allebach, Jan P.

doi:10.1109/icip.2019.8803063

Cited by 4 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(Liang et al, 2022) introduces Multidomain TriSeNet Networks for the real-time single person segmentation for photo-editing applications. (Xi et al, 2019) uses saliency map derived from accurate pose information to improve segmentation accuracy especially in multi-person scenes. (Han et al, 2020) categorizes the set of techniques for reducing model size into model compression and compact model design methods.…”

Section: Related Workmentioning

confidence: 99%

ALiSNet: Accurate and Lightweight Human Segmentation Network for Fashion E-Commerce

Seifoddini¹,

Vernooij²,

Timon³

et al. 2023

Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

Accurately estimating human body shape from photos can enable innovative applications in fashion, from mass customization, to size and fit recommendations and virtual try-on. Body silhouettes calculated from user pictures are effective representations of the body shape for downstream tasks. Smartphones provide a convenient way for users to capture images of their body, and on-device image processing allows predicting body segmentation while protecting users' privacy. Existing off-the-shelf methods for human segmentation are closed source and cannot be specialized for our application of body shape and measurement estimation. Therefore, we create a new segmentation model by simplifying Semantic FPN with PointRend, an existing accurate model. We finetune this model on a high-quality dataset of humans in a restricted set of poses relevant for our application. We obtain our final model, ALiSNet, with a size of 4MB and 97.6 ± 1.0% mIoU, compared to Apple Person Segmentation, which has an accuracy of 94.4 ± 5.7% mIoU on our dataset.

show abstract

Section: Related Workmentioning

confidence: 99%