Where Are My Clothes? a Multi-Level Approach for Evaluating Deep Instance Segmentation Architectures on Fashion Images

Warren Jouanneau, Aurélie Bugeau, Marc Palyart, Nicolas Papadakis, Laurent Vézard

CVPRW 2021 pp. 3951-3955

doi:10.1109/CVPRW53098.2021.00443 /cvprw/2021/jouanneau2021cvprw-my/

Abstract

In this paper we present an extensive evaluation of instance segmentation in the context of images containing clothes. We propose a multi level evaluation that completes the classical overlapping criteria given by IoU. In particular, we quantify both the contour and color content accuracy of the the predicted segmentation masks. We demonstrate that the proposed evaluation framework is relevant to obtain meaningful insights on models performance through experiments conducted on five state of the art instance segmentation methods.

PDF CVPRW Semantic Scholar

Cite

Text

Jouanneau et al. "Where Are My Clothes? a Multi-Level Approach for Evaluating Deep Instance Segmentation Architectures on Fashion Images." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2021. doi:10.1109/CVPRW53098.2021.00443

Markdown

[Jouanneau et al. "Where Are My Clothes? a Multi-Level Approach for Evaluating Deep Instance Segmentation Architectures on Fashion Images." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2021.](https://mlanthology.org/cvprw/2021/jouanneau2021cvprw-my/) doi:10.1109/CVPRW53098.2021.00443

BibTeX

@inproceedings{jouanneau2021cvprw-my,
  title     = {{Where Are My Clothes? a Multi-Level Approach for Evaluating Deep Instance Segmentation Architectures on Fashion Images}},
  author    = {Jouanneau, Warren and Bugeau, Aurélie and Palyart, Marc and Papadakis, Nicolas and Vézard, Laurent},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2021},
  pages     = {3951-3955},
  doi       = {10.1109/CVPRW53098.2021.00443},
  url       = {https://mlanthology.org/cvprw/2021/jouanneau2021cvprw-my/}
}