Yizhao Wang

Yizhao Wang

Auto-generated profile, based on publicly available information

Yizhao Wang

School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai, China

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
12 December 2023 in Journal of Manufacturing Systems
Verified
Journal Article
Closed access
ZZZhuangzhuang Zhang
Zhuangzhuang Zhang
YWYizhao Wang
Yizhao Wang
ZZZhinan Zhang
Zhinan Zhang
LWLihui Wang
Lihui Wang
HHHuang Huang
Huang Huang
QCQixin Cao
Qixin Cao

17 April 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
HWHaili Wang
Haili Wang
FDFuchuan Du
Fuchuan Du
HZHao Zhu
Hao Zhu
ZZZhuangzhuang Zhang
Zhuangzhuang Zhang
YWYizhao Wang
Yizhao Wang
XZXiaoxiao Zhu
Xiaoxiao Zhu
The non-contact monitoring of vital signs without body-attached sensors is a promising home healthcare technique. Recent studies have shown that millimeter-wave (mmW) radar can be used to continuous daily monitoring of vital signs such as heart rate (HR) and respiratory rate (RR). To minimize disruption to the user’s daily life and accommodate for home environments, low-power radar or radar with antennas in package (AiP) is becoming increasingly popular. However, the lower transmitting power of these systems can lead to decreased Signal-to-Noise Ratio (SNR) of the radar signal, making established measurement methods ineffective. The vital signs, particularly HR, are no longer accurately measured due to the small amplitude associated with heartbeats. In this work, we present HeRe for low-power radar, a deep learning approach for heartbeat signal reconstruction. The proposed method utilizes a neural network to identify patterns in the signal, leading to a significant improvement in the SNR of the heartbeat signal. By leveraging historical data, the current HR estimation is further refined, yielding improved precision in the results. Experiments have demonstrated the efficacy of signal reconstruction for improving the heartbeat signal in both the time and frequency domains. A comparison of the proposed approach to related works on various HR distributions, sensing ranges, and subjects showed significantly higher accuracy. To further assess the robustness of the proposed approach, extended continuous monitoring was performed and 97.5% accuracy was achieved. This article highlights the advantages of deep learning techniques for non-contact, radar-based HR monitoring, leading to accurate HR monitoring with the lowest power consumer-grade millimeter wave radar currently available on the market.

25 December 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
HSHan Sun
Han Sun
ZZZhuangzhuang Zhang
Zhuangzhuang Zhang
HWHaili Wang
Haili Wang
YWYizhao Wang
Yizhao Wang
QCQixin Cao
Qixin Cao
Industrial bin picking is a challenging task as lots of highly reflective and textureless objects are stacked. Previous methods generally predict gripper configurations or 6D object pose by expensive industrial 3D camera, ignoring the role of consumer-grade RGB-D cameras in this task. For highly reflective objects, depth maps of consumer-grade RGB-D cameras are generally of a lower quality with much stronger noise, making it challenging to detect robust grasp poses in stacked industrial scenarios. To address these challenges, we propose a novel grasp detection framework including a synthetic data generation pipeline, a novel local grasp detection method, and a global search algorithm. Specifically, in the training phase, we utilize the Pybullet and OpenGL to generate synthetic data by approximate 3D models. We introduce a novel grasp representation and propose MixStyle-ResUnet to predict and achieve sim2real transfer for this representation. In the testing phase, the Object Singleness Metric (OSM) is proposed for the global search block to evaluate the graspable probability. In addition, the RGB image enhanced by edge feature is utilized as the input of the network, which can bridge the domain shift. Extensive experiments in real scenes demonstrate that our method can achieve competitive results in grasping disordered metal parts. In over 2400 robotic grasp trials, our method achieves an average success rate of 91.1% in dense stack scenarios. We also confirm that the proposed method can be applied to unseen objects which are not included in the training dataset. Code and videos are available at https://github.com/sunhan1997/IndusGrasp.

22 March 2024 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
HSHan Sun
Han Sun
ZZZhenning Zhou
Zhenning Zhou
YWYizhao Wang
Yizhao Wang
ZZZhuangzhuang Zhang
Zhuangzhuang Zhang
QCQixin Cao
Qixin Cao
The 6D pose estimation for metal parts is essential in industrial robotic applications. The color homogeneity, texture-less and light-reflecting properties of metal parts raise great challenges. Current 6D pose estimation methods have gained extensive concern using CNNs. However, these CNN-based methods lack Transformer's ability to focus on extracting low-frequency features and long-range context information. In the study, we explore taking full advantage of CNN and Transformer from a frequency-domain perspective to enhance the performance of metal parts' 6D pose estimation. Specifically, we propose a frequency-guided CNN-Transformer fusion 6D pose estimation network (FGCT6D). First, we construct a novel pixel attention residual module to improve the high-frequency attention of CNN. Then, we design a dual-branch CNN-Transformer encoder: the Swin-Transformer extracts global information and low-frequency features, and the CNN captures local information and high-frequency features. Second, the frequency-guided feature fusion module is proposed to fuse the extracted multi-spectral features. Third, to maximize the utilization of the rich frequency-domain feature representation, we propose a feature fusion decoder with Conv-MSA modules. Additionally, we leverage optimal transport theory, treating dense correspondences as spatial probability distributions, and design the optimal transport loss function. Experiments show that our method can extract rich frequency-domain features, and achieve competitive performance on the MP6D and LINEMOD datasets.

26 May 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
HSHan Sun
Han Sun
YWYizhao Wang
Yizhao Wang
ZZZhenning Zhou
Zhenning Zhou
MLMingyang Li
Mingyang Li
NLNailong Liu
Nailong Liu
QCQixin Cao
Qixin Cao
The 6D pose estimation for metal parts is essential in industrial robotic applications. Although the 6D pose estimation methods that rely on object-specific training have gained extensive concern, these methods can’t generalize to novel objects. Recent novel object pose estimation methods are solving this issue using task-specific fine-tuned CNNs for deep template matching. However, these methods require expensive training procedures and don’t consider pose distribution of rendered templates. Recently, foundation models show strong representation learning ability, and can encode both the high spatial information as well as semantic information. In this study, we present a metal parts’ zero-shot 6D pose estimation method by foundation model without re-training, and incorporate prior information of metal parts’ poses to generate rendered templates that align with the pose distribution in real world. DINOv2, a recent vision foundation model with impressive generalization capabilities, is employed for matching rendered templates against query images of metal parts. Additionally, we introduce the optimal transport as a similarity metric. Then, the Gluestick is utilized to establish local keypoint correspondences, which enable deriving geometric correspondences and are used for estimating the metal part’s 6D pose with PnP/RANSAC. Given an industrial scenario, we first estimate metal parts’ pose, select and save high-confidence pose results to pose buffer, which is used to update the templates. The experiments on MP6D and ROBI datasets showcase that the proposed method has better performance than MegaPose. We also conduct real world experiments, which demonstrate the robustness of the proposed method. Code is available at https://github.com/sunhan1997/IndusPose.

3 January 2021 in International Journal of Computer Assisted Radiology and Surgery
Verified
Journal Article
Closed access
PDF file
YWYizhao Wang
Yizhao Wang
QCQixin Cao
Qixin Cao
XZXiaoxiao Zhu
Xiaoxiao Zhu
PWPengfei Wang
Pengfei Wang