Localization and Completion for 3D Object Interactions
作者： Xi Zhao , Ruizhen Hu , Haisong Liu , Taku Komura , Xinyu Yang
DOI: 10.1109/TVCG.2019.2892454
关键词:
摘要: Finding where and what objects to put into an existing scene is a common task for synthesis robot/character motion planning. Existing frameworks require development of hand-crafted features suitable the task, or full volumetric analysis that could be memory intensive imprecise. In this paper, we propose data-driven framework discover location then place appropriate in scene. Our approach inspired by computer vision techniques localizing images: using all directional depth image (ADD-image) encodes 360-degree field view from samples scene, our system regresses images positions new object can located. Given several candidate areas around host predicts partner whose geometry fits well object. highly parallel efficient, especially handling interactions between large small objects. We show examples hang bags on hooks, fit chairs front desks, shelves, insert flowers vases, hangers onto laundry rack.
computer.org
ed.ac.uk
sci-hub.se 参考文章(23)
Shaoqing Ren, Kaiming He, Ross Girshick, Jian Sun, Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 39, pp. 1137- 1149 ,(2017) , 10.1109/TPAMI.2016.2577031
Ross Girshick, Fast R-CNN international conference on computer vision. pp. 1440- 1448 ,(2015) , 10.1109/ICCV.2015.169
Joseph Redmon, Anelia Angelova, Real-time grasp detection using convolutional neural networks international conference on robotics and automation. pp. 1316- 1322 ,(2015) , 10.1109/ICRA.2015.7139361
Zhirong Wu, Shuran Song, Aditya Khosla, Fisher Yu, Linguang Zhang, Xiaoou Tang, Jianxiong Xiao, 3D ShapeNets: A deep representation for volumetric shapes computer vision and pattern recognition. pp. 1912- 1920 ,(2015) , 10.1109/CVPR.2015.7298801
Xi Zhao, He Wang, Taku Komura, Indexing 3D Scenes Using the Interaction Bisector Surface ACM Transactions on Graphics. ,vol. 33, pp. 22- ,(2014) , 10.1145/2574860
Lucas Majerowicz, Ariel Shamir, Alla Sheffer, Holger H. Hoos, Filling Your Shelves: Synthesizing Diverse Style-Preserving Artifact Arrangements. IEEE Transactions on Visualization and Computer Graphics. ,vol. 20, pp. 1507- 1518 ,(2014) , 10.1109/TVCG.2013.245
Ruizhen Hu, Chenyang Zhu, Oliver van Kaick, Ligang Liu, Ariel Shamir, Hao Zhang, Interaction context (ICON): towards a geometric functionality descriptor international conference on computer graphics and interactive techniques. ,vol. 34, pp. 83- ,(2015) , 10.1145/2766914
Matthew Fisher, Manolis Savva, Pat Hanrahan, Characterizing structural relationships in scenes using graph kernels ACM SIGGRAPH 2011 papers on - SIGGRAPH '11. ,vol. 30, pp. 34- ,(2011) , 10.1145/1964921.1964929
Kang Chen, Yu-Kun Lai, Yu-Xin Wu, Ralph Martin, Shi-Min Hu, Automatic semantic modeling of indoor scenes from low-quality RGB-D data using contextual information international conference on computer graphics and interactive techniques. ,vol. 33, pp. 208- ,(2014) , 10.1145/2661229.2661239
Ross Girshick, Jeff Donahue, Trevor Darrell, Jitendra Malik, Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation computer vision and pattern recognition. pp. 580- 587 ,(2014) , 10.1109/CVPR.2014.81