no code implementations • 3 Apr 2024 • Xu Wang, YiFan Li, Qiudan Zhang, Wenhui Wu, Mark Junjie Li, Jianmin Jinag
However, previous 3D scene graph generation methods utilize a fully supervised learning manner and require a large amount of entity-level annotation data of objects and relations, which is extremely resource-consuming and tedious to obtain.
no code implementations • 20 Sep 2022 • Haohong Liao, Silin Zheng, Xuelin Shen, Mark Junjie Li, Xu Wang
However, according to our investigation, the lacking of datasets focusing on real-world application scenarios limits the further improvements for current learning-based MTMCT models.
no code implementations • 6 Jun 2019 • Zhalama, Jiji Zhang, Frederick Eberhardt, Wolfgang Mayer, Mark Junjie Li
In recent years the possibility of relaxing the so-called Faithfulness assumption in automated causal discovery has been investigated.