Hongyu Wu

Yinglong Li, Hongyu Wu, Xiaogang Wang et al.

We propose FaceCom, a method for 3D facial shape completion, which delivers high-fidelity results for incomplete facial inputs of arbitrary forms. Unlike end-to-end shape completion methods based on point clouds or voxels, our approach relies on a mesh-based generative network that is easy to optimize, enabling it to handle shape completion for irregular facial scans. We first train a shape generator on a mixed 3D facial dataset containing 2405 identities. Based on the incomplete facial input, we fit complete faces using an optimization approach under image inpainting guidance. The completion results are refined through a post-processing step. FaceCom demonstrates the ability to effectively and naturally complete facial scan data with varying missing regions and degrees of missing areas. Our method can be used in medical prosthetic fabrication and the registration of deficient scanning data. Our experimental results demonstrate that FaceCom achieves exceptional performance in fitting and shape completion tasks. The code is available at https://github.com/dragonylee/FaceCom.git.

Xuhui Fan, Zhangkai Wu, Hongyu Wu

Diffusion Models~(DMs) have emerged as the dominant approach in Generative Artificial Intelligence (GenAI), owing to their remarkable performance in tasks such as text-to-image synthesis. However, practical DMs, such as stable diffusion, are typically trained on massive datasets and thus usually require large storage. At the same time, many steps may be required, i.e., recursively evaluating the trained neural network, to generate a high-quality image, which results in significant computational costs during sample generation. As a result, distillation methods on pre-trained DM have become widely adopted practices to develop smaller, more efficient models capable of rapid, few-step generation in low-resource environment. When these distillation methods are developed from different perspectives, there is an urgent need for a systematic survey, particularly from a methodological perspective. In this survey, we review distillation methods through three aspects: output loss distillation, trajectory distillation and adversarial distillation. We also discuss current challenges and outline future research directions in the conclusion.

Zhangkai Wu, Xuhui Fan, Hongyu Wu et al.

Pre-trained diffusion models are commonly used to generate clean data (e.g., images) from random noises, effectively forming pairs of noises and corresponding clean images. Distillation on these pre-trained models can be viewed as the process of constructing advanced trajectories within the pair to accelerate sampling. For instance, consistency model distillation develops consistent projection functions to regulate trajectories, although sampling efficiency remains a concern. Rectified flow method enforces straight trajectories to enable faster sampling, yet relies on numerical ODE solvers, which may introduce approximation errors. In this work, we bridge the gap between the consistency model and the rectified flow method by proposing a Straight Consistent Trajectory~(SCoT) model. SCoT enjoys the benefits of both approaches for fast sampling, producing trajectories with consistent and straight properties simultaneously. These dual properties are strategically balanced by targeting two critical objectives: (1) regulating the gradient of SCoT's mapping to a constant, (2) ensuring trajectory consistency. Extensive experimental results demonstrate the effectiveness and efficiency of SCoT.

A. Khaled Zarabie, Sanjoy Das, Hongyu Wu

While non-parametric models, such as neural networks, are sufficient in the load forecasting, separate estimates of fixed and shiftable loads are beneficial to a wide range of applications such as distribution system operational planning, load scheduling, energy trading, and utility demand response programs. A semi-parametric estimation model is usually required, where cost sensitivities of demands must be known. Existing research work consistently uses somewhat arbitrary parameters that seem to work best. In this paper, we propose a generic class of data-driven semiparametric models derived from consumption data of residential consumers. A two-stage machine learning approach is developed. In the first stage, disaggregation of the load into fixed and shiftable components is accomplished by means of a hybrid algorithm consisting of non-negative matrix factorization (NMF) and Gaussian mixture models (GMM), with the latter trained by an expectation-maximization (EM) algorithm. The fixed and shiftable loads are subject to analytic treatment with economic considerations. In the second stage, the model parameters are estimated using an L2-norm, epsilon-insensitive regression approach. Actual energy usage data of two residential customers show the validity of the proposed method.

Bo Liu, Hongyu Wu, Yingchen Zhang et al.

Due to the insufficient measurements in the distribution system state estimation (DSSE), full observability and redundant measurements are difficult to achieve without using the pseudo measurements. The matrix completion state estimation (MCSE) combines the matrix completion and power system model to estimate voltage by exploring the low-rank characteristics of the matrix. This paper proposes a robust matrix completion state estimation (RMCSE) to estimate the voltage in a distribution system under a low-observability condition. Tradition state estimation weighted least squares (WLS) method requires full observability to calculate the states and needs redundant measurements to proceed a bad data detection. The proposed method improves the robustness of the MCSE to bad data by minimizing the rank of the matrix and measurements residual with different weights. It can estimate the system state in a low-observability system and has robust estimates without the bad data detection process in the face of multiple bad data. The method is numerically evaluated on the IEEE 33-node radial distribution system. The estimation performance and robustness of RMCSE are compared with the WLS with the largest normalized residual bad data identification (WLS-LNR), and the MCSE.

M. Nazif Faqiry, Lawryn Edmonds, Hongyu Wu

The largescale penetration of variable renewable energy (VRE) and their generation uncertainties poses a major challenge for the distribution system operator (DSO) to efficiently determine the day-ahead real and reactive power distribution locational marginal prices (DLMPs) and their underlying components. In this paper, we propose a DLMP-based transactive day-ahead market (DAM) model, that in addition to energy and losses, determines prices for creating congestions and voltage violations under peak-load and large-scale stochastic VRE penetration conditions. To account for the VRE uncertainties and the effect of their large-scale penetration on the DLMP components and distributed energy resources' (DERs) schedules, we propose a novel data-driven probability efficient point (PEP) method that computes the optimal total VRE generation at different confidence (risk) levels to incorporate in the proposed transactive DAM model. We perform a wide range of simulation studies on a modified IEEE 69-node system to validate the proposed methods and demonstrate the effect of peak load conditions, large-scale VRE penetration, and inclusion of battery energy storage systems (BESS) on the resulting positive or negative real and reactive power DLMPs and their components.