Z
Zhongang Cai
Researcher at SenseTime
Publications - 34
Citations - 412
Zhongang Cai is an academic researcher from SenseTime. The author has contributed to research in topics: Computer science & Deep learning. The author has an hindex of 4, co-authored 22 publications receiving 57 citations. Previous affiliations of Zhongang Cai include Nanyang Technological University.
Papers
More filters
Proceedings ArticleDOI
Variational Relational Point Completion Network
TL;DR: Li et al. as mentioned in this paper proposed a dual-path architecture to enable principled probabilistic modeling across partial and complete clouds, where one path consumes complete point clouds for reconstruction by learning a point VAE and the other path generates complete shapes for partial point clouds, whose embedded distribution is guided by distribution obtained from the reconstruction path during training.
Journal ArticleDOI
AvatarCLIP: Zero-Shot Text-Driven Generation and Animation of 3D Avatars
TL;DR: By leveraging the priors learned in the motion VAE, a CLIP-guided reference-based motion synthesis method is proposed for the animation of the generated 3D avatar, which validate the effectiveness and generalizability of texture generation.
Proceedings ArticleDOI
Unsupervised 3D Shape Completion through GAN Inversion
Junzhe Zhang,Xinyi Chen,Zhongang Cai,Liang Pan,Haiyu Zhao,Shuai Yi,Chai Kiat Yeo,Bo Dai,Chen Change Loy +8 more
TL;DR: ShapeInversion as mentioned in this paper uses a GAN pre-trained on complete shapes by searching for a latent code that gives a complete shape that best reconstructs the given partial input, which is capable of incorporating the rich prior captured in a well-trained generative model.
Proceedings ArticleDOI
Siamese Convolutional Neural Network for Sub-millimeter-accurate Camera Pose Estimation and Visual Servoing
TL;DR: A new neural network, based on a Siamese architecture, is proposed, for highly accurate camera pose estimation, that can generalize to similar objects, is robust against changing lighting conditions, and to partial occlusions (when used iteratively).
Proceedings ArticleDOI
HuMMan: Multi-Modal 4D Human Dataset for Versatile Sensing and Modeling
Zhongang Cai,Daxuan Ren,Ailing Zeng,Zhengyu Lin,Wenjia Wang,Xiangyu Fan,Yangmin Gao,Yifan Yu,Liang Pan,Fangzhou Hong,Mingyuan Zhang,Chen Change Loy,Lei Juan Yang,Ziwei Liu +13 more
TL;DR: HuMMan is a large-scale multi-modal 4D human dataset with 1000 human subjects, 400k sequences and 60M frames that voice the need for further study on challenges such as Ne-grained action recognition, dynamic human mesh reconstruction, and textured mesh reconstruction.