Active Object Reconstruction Using a Guided View Planner

Abstract

Inspired by the recent advance of image-based object reconstruction using deep learning, we present an active reconstruction model using a guided view planner. We aim to reconstruct a 3D model using images observed from a planned sequence of informative and discriminative views. But where are such informative and discriminative views around an object? To address this we propose a unified model for view planning and object reconstruction, which is utilized to learn a guided information acquisition model and to aggregate information from a sequence of images for reconstruction. Experiments show that our model (1) increases our reconstruction accuracy with an increasing number of views (2) and generally predicts a more informative sequence of views for object reconstruction compared to other alternative methods.

Cite

Text

Yang et al. "Active Object Reconstruction Using a Guided View Planner." International Joint Conference on Artificial Intelligence, 2018. doi:10.24963/IJCAI.2018/689

Markdown

[Yang et al. "Active Object Reconstruction Using a Guided View Planner." International Joint Conference on Artificial Intelligence, 2018.](https://mlanthology.org/ijcai/2018/yang2018ijcai-active/) doi:10.24963/IJCAI.2018/689

BibTeX

@inproceedings{yang2018ijcai-active,
  title     = {{Active Object Reconstruction Using a Guided View Planner}},
  author    = {Yang, Xin and Wang, Yuanbo and Wang, Yaru and Yin, Baocai and Zhang, Qiang and Wei, Xiaopeng and Fu, Hongbo},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2018},
  pages     = {4965-4971},
  doi       = {10.24963/IJCAI.2018/689},
  url       = {https://mlanthology.org/ijcai/2018/yang2018ijcai-active/}
}