Mi Zhou

For now, I am a PhD student at the Shenzhen International Graduate School, Tsinghua University, where I work on holographic display for virtual and augmented reality. My PhD advisor is Zihan Geng. Before that, I was a undergraduate student of Central South University (CSU) majoring on communication engineering.

zhoumicsu[at]qq.com  /  GitHub  /  Google Scholar

Research

I'm interested in computational display, imaging, VR, and AR.

Physics-aware cross-domain fusion aids learning-driven computer-generated holography

Speckle-free holography with a diffraction-aware global perceptual model

we propose a CGH model called Holomer. Its single-layer perceptual field is 43 times larger than that of a widely used convolutional kernel, thanks to the embedding-based feature dimensionality reduction and multi-head sliding-window self-attention mechanisms.

Prior-free 3D tracking of a fast-moving object at 6667 frames per second with single-pixel detectors

We propose a novel approach that enables real-time 3D tracking of a fast-moving object without any prior motion information and at a very low computational cost.

Error-compensation network for ringing artifact reduction in holographic displays

We explore a diffraction propagation error-compensation network that can be easily integrated into existing CGH methods. This network is designed to correct propagation errors by predicting residual values, thereby aligning the diffraction process closely with an ideal state and easing the learning burden of the network.

Adaptive super-resolution networks for single-pixel imaging at ultra-low sampling rates

We introduce a network architecture specifically tailored for SPI, which demonstrates improved performance even before integrating with SPI’s physical sampling processes. This integration, particularly focusing on the nuanced effects of sampling rates within the model’s loss function and data preprocessing, enhances image reconstruction quality and adaptability at low sampling rates, down to 1.56%.

Point spread function-inspired deformable convolutional network for holographic displays

We propose deformable holography (DeH) algorithm for CGH. We demonstrate that utilizing deformable convolutions enable adaptive modeling of geometric transformations.

Image-free single-pixel keypoint detection for privacy preserving human pose estimation

Our proposed method can complete the keypoint detection task at an ultralow sampling rate on a measured one-dimensional sequence without image reconstruction, thus protecting privacy from the data collection stage and preventing the acquisition of detailed visual information from the source.

End-to-end compression-aware computer-generated holography

We introduce a novel approach to merge the process of hologram generation and JPEG compression with one differentiable model, enabling joint optimization via efficient first-order solvers.

A wireless integrated system with hybrid embedded sensing for the continuous monitoring of bird flight

The proposed system leverages a strain sensor based on laser-induced graphene (LIG), strategically positioned on the bird’s wing joints, in conjunction with a high-precision IMU deployed on the bird’s torso. Through a learning architecture, we achieve an impressive accuracy of 99.48% in identifying eight commonly observed bird flight postures.

Ultra-efficient single-pixel tracking and imaging of moving objects based on geometric moment

We propose a novel method that only needs to reconstruct one image of the object during the whole moving process, thus significantly reducing the computational cost. We achieve ultra-efficient simultaneously tracking and imaging of a moving object by designing novel illumination patterns.

GAN-SRSPI: super-resolution single-pixel imaging using generative adversarial networks

This work proposes a super-resolution single-pixel imaging methodology based on generative adversarial networks (GAN-SRSPI). A low-resolution (N×N) image is reconstructed and then super-resolved to obtain a high-resolution (4N× 4N) image.

A digital twin model for battery management systems: concepts, algorithms, and platforms

We first discuss the corresponding concepts about the digital twin model of battery management systems. Then, the state-of-charge (SoC) and state-of-health (SoH) estimation algorithms are presented in an integrated fashion for the monitoring and prognostics. Concretely, the extended Kalman filter algorithm (EKF) is used in this paper for the estimation of SoC, which improves the robustness of digital twin model, and the particle swarm optimization algorithm (PSO) is used in this paper for the estimation of SoH.

Design and source code from Jon Barron's website