Method | Score 1 | Score 2 | Score 3 | Score 4 | Submission Time | Contributors | Description |
---|---|---|---|---|---|---|---|
URNet | 0.6017 | 0.5511 | 0.4229 | 0.5252 | Oct. 17, 2019, 6:30 p.m. | Tianfei Zhou, Wenguan Wang, Zhijie Zhang, Jianbing Shen (Inception Institute of Artificial Intelligence, Tianjin University) | We propose a Unified Relationship detection Network (URNet) in this contest. We jointly train entity detection/segmentation and relationship inference in an end-to-end way. Our URNet is built on Cascade (Mask)RCNN, which is further extended to add a new branch for relationship recognition for each human-object pair. The relationship recognition branch operates on rich multi-modal features for accurate inference. |
HTC+iCAN | 0.5621 | 0.5232 | 0.3749 | 0.4867 | Oct. 16, 2019, 5:22 p.m. | Wenbo Song (Tianjin University) | We combined Hybrid Task Cascade with iCAN to generate the scene graph. |
RelNet | 0.5317 | 0.4926 | 0.3244 | 0.4496 | Oct. 17, 2019, 3:49 p.m. | Kai Chen, Chao Xu, Jiahui Li (Sun Yat-sen University) | RelNet |
XNet | 0.3842 | 0.3315 | 0.1729 | 0.2962 | Oct. 12, 2019, 11:21 a.m. | – | – |
AI is evolving. Don't get left behind.
AI insights delivered straight to your inbox.