Submitted by Heng Wang.

Submission data

Full nameSpatiality-guided Transformer for 3D Dense Captioning on Point Clouds
DescriptionVoteNet + spatiality-guided Transformer
Publication titleSpatiality-guided Transformer for 3D Dense Captioning on Point Clouds
Publication authorsHeng Wang, Chaoyi Zhang, Jianhui Yu, Weidong Cai
Publication venuethe 31st International Joint Conference on Artificial Intelligence (IJCAI), 2022
Publication URLhttps://www.ijcai.org/proceedings/2022/0194.pdf
Input Data TypesUses XYZ coordinates,Uses RGB values,Uses Normal Vectors
Programming language(s)Python
HardwareGeForce RTX 2080 Ti
Websitehttps://spacap3d.github.io/
Source code or download URLhttps://github.com/heng-hw/SpaCap3D
Submission creation date31 Aug, 2022
Last edited31 Aug, 2022

Captioning

Captioning F1-Score Dense Captioning Object Detection
CIDEr@0.5IoUBLEU-4@0.5IoURouge-L@0.5IoUMETEOR@0.5IoUDCmAPmAP@0.5
0.13590.08830.15910.07380.11820.3275