Amodal Instance Segmentation via Prior-Guided Expansion

Junjie Chen, Li Niu, Jianfu Zhang, Jianlou Si, Chen Qian, Liqing Zhang

AAAI 2023 pp. 313-321

doi:10.1609/AAAI.V37I1.25104 /aaai/2023/chen2023aaai-amodal/

Abstract

Amodal instance segmentation aims to infer the amodal mask, including both the visible part and occluded part of each object instance. Predicting the occluded parts is challenging. Existing methods often produce incomplete amodal boxes and amodal masks, probably due to lacking visual evidences to expand the boxes and masks. To this end, we propose a prior-guided expansion framework, which builds on a two-stage segmentation model (i.e., Mask R-CNN) and performs box-level (resp., pixel-level) expansion for amodal box (resp., mask) prediction, by retrieving regression (resp., flow) transformations from a memory bank of expansion prior. We conduct extensive experiments on KINS, D2SA, and COCOA cls datasets, which show the effectiveness of our method.

PDF AAAI Semantic Scholar

Cite

Text

Chen et al. "Amodal Instance Segmentation via Prior-Guided Expansion." AAAI Conference on Artificial Intelligence, 2023. doi:10.1609/AAAI.V37I1.25104

Markdown

[Chen et al. "Amodal Instance Segmentation via Prior-Guided Expansion." AAAI Conference on Artificial Intelligence, 2023.](https://mlanthology.org/aaai/2023/chen2023aaai-amodal/) doi:10.1609/AAAI.V37I1.25104

BibTeX

@inproceedings{chen2023aaai-amodal,
  title     = {{Amodal Instance Segmentation via Prior-Guided Expansion}},
  author    = {Chen, Junjie and Niu, Li and Zhang, Jianfu and Si, Jianlou and Qian, Chen and Zhang, Liqing},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2023},
  pages     = {313-321},
  doi       = {10.1609/AAAI.V37I1.25104},
  url       = {https://mlanthology.org/aaai/2023/chen2023aaai-amodal/}
}