Guangyuan Wang

Guangyuan Wang

Auto-generated profile, based on publicly available information

Guangyuan Wang

Tongyi Lab, Alibaba, Hangzhou, China

Research Focus & Metrics

Academic Career Length

17Years (2010-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 August 2023 in ACM Transactions on Graphics
Verified
Journal Article
Open access
STSicong Tang
Sicong Tang
GWGuangyuan Wang
Guangyuan Wang
QRQing Ran
Qing Ran
LLLingzhi Li
Lingzhi Li
LSLi Shen
Li Shen
PTPing Tan
Ping Tan
We present a novel method for reconstructing clothed humans from a sparse set of, e.g., 1–6 RGB images. Despite impressive results from recent works employing deep implicit representation, we revisit the volumetric approach and demonstrate that better performance can be achieved with proper system design. The volumetric representation offers significant advantages in leveraging 3D spatial context through 3D convolutions, and the notorious quantization error is largely negligible with a reasonably large yet affordable volume resolution, e.g., 512. To handle memory and computation costs, we propose a sophisticated coarse-to-fine strategy with voxel culling and subspace sparse convolution. Our method starts with a discretized visual hull to compute a coarse shape and then focuses on a narrow band nearby the coarse shape for refinement. Once the shape is reconstructed, we adopt an image-based rendering approach, which computes the colors of surface points by blending input images with learned weights. Extensive experimental results show that our method significantly reduces the mean point-to-surface (P2S) precision of state-of-the-art methods by more than 50% to achieve approximately 2mm accuracy with a 512 volume resolution. Additionally, images rendered from our textured model achieve a higher peak signal-to-noise ratio (PSNR) compared to state-of-the-art methods.

1 August 2010
Verified
Conference Paper
Closed access
HWHao Wang
Hao Wang
GTGuozhen Tan
Guozhen Tan
YWYuan Wang
Yuan Wang
GWGuangyuan Wang
Guangyuan Wang
The concept of the Internet of Things provides a new model to access the information of the objective world. A new approach based on the Internet of Things for traffic emergent events is proposed. The network architecture entails a hierarchy of capability, information and control, where nodes in the network expected to possess resources for networking and computing and presume autonomy through multi-functional modules for sensory processing and situation assessment. For the data fusion of multi-sensor device, the paper puts forward three different methods of date fusion according to different integrality levels of raw information. Finally, Bayesian network method is used to get situational assessment of the fused data. Comprehensive experiments on urban traffic emergent events of Dalian and comparisons with several other methods show that the Bayesian network combined with the Internet of Things is a very promising and effective approach for traffic emergent events' situation assessment modeling and forecasting, both for complete data and incomplete data.