metadata

license: cc-by-nc-4.0
task_categories:
  - image-segmentation
tags:
  - matting
  - instance matting
  - image matting
  - video matting
  - guidance matting
  - human matting
pretty_name: MaGGIe - Human Instance Image and Video Matting

MaGGIe: Mask Guided Gradual Human Instance Matting

[Project Page] [Code]

Train datasets and Benchmarks for Instance-awareness alpha human matting with binary mask guidance for images and video

Accepted at CVPR 2024

Chuong Huynh, Seoung Wug Oh, Abhinav Shrivastava, Joon-Young Lee

Work is a part of Summer Internship 2023 at Adobe Research

Please refer to our paper for details.

Citation

If you find MaGGIe useful in your research, please cite the following paper:

@article{chuonghm_maggie,
      author    = {Chuong Huynh and Seoung Wug Oh and and Abhinav Shrivastava and Joon-Young Lee},
      title     = {MaGGIe: Masked Guided Gradual Human Instance Matting},
      journal   = {arXiv:2404.16035},
      year      = {2024}
}