TY - GEN
T1 - Neural Volumetric Object Selection
AU - Ren, Zhongzheng
AU - Agarwala, Aseem
AU - Russell, Bryan
AU - Schwing, Alexander G.
AU - Wang, Oliver
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - We introduce an approach for selecting objects in neural volumetric 3D representations, such as multi-plane images (MPI) and neural radiance fields (NeRF). Our approach takes a set of foreground and background 2D user scribbles in one view and automatically estimates a 3D segmentation of the desired object, which can be rendered into novel views. To achieve this result, we propose a novel voxel feature embedding that incorporates the neural volumetric 3D representation and multi-view image features from all input views. To evaluate our approach, we introduce a new dataset of human-provided segmentation masks for depicted objects in real-world multi-view scene captures. We show that our approach out-performs strong baselines, including 2D segmentation and 3D segmentation approaches adapted to our task.
AB - We introduce an approach for selecting objects in neural volumetric 3D representations, such as multi-plane images (MPI) and neural radiance fields (NeRF). Our approach takes a set of foreground and background 2D user scribbles in one view and automatically estimates a 3D segmentation of the desired object, which can be rendered into novel views. To achieve this result, we propose a novel voxel feature embedding that incorporates the neural volumetric 3D representation and multi-view image features from all input views. To evaluate our approach, we introduce a new dataset of human-provided segmentation masks for depicted objects in real-world multi-view scene captures. We show that our approach out-performs strong baselines, including 2D segmentation and 3D segmentation approaches adapted to our task.
KW - 3D from multi-view and sensors
KW - 3D from single images
KW - Segmentation
KW - grouping and shape analysis
UR - http://www.scopus.com/inward/record.url?scp=85138296073&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85138296073&partnerID=8YFLogxK
U2 - 10.1109/CVPR52688.2022.00604
DO - 10.1109/CVPR52688.2022.00604
M3 - Conference contribution
AN - SCOPUS:85138296073
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 6123
EP - 6132
BT - Proceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
PB - IEEE Computer Society
T2 - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
Y2 - 19 June 2022 through 24 June 2022
ER -