Kecheng Shi

Kecheng Shi

Auto-generated profile, based on publicly available information

Kecheng Shi

School of Automation Engineering, University of Electronic Science and Technology of China, Chengdu, 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
24 April 2025 in IEEE Transactions on Automation Science and Engineering
Verified
Journal Article
Closed access
KSKecheng Shi
Kecheng Shi
RHRui Huang
Rui Huang
JLJianzhi Lyu
Jianzhi Lyu
ZLZhe Li
Zhe Li
FMFengjun Mu
Fengjun Mu
BGBijoy Ghosh
Bijoy Ghosh
Motor imagery-based brain–computer interfaces (MI-BCIs) hold significant promise for rehabilitation training in individuals with neurological impairments such as stroke and spinal cord injury (SCI). Achieving precise and robust lower limb movement prediction for each patient is crucial. However, the variability in MI response frequencies and brain activation patterns among subjects presents a great challenge to the generalizability of MI-BCIs. This paper proposes a Tuned Heuristic Fusion Graph Convolutional Network (THFGCN) for limb movement prediction in rehabilitation scenarios. THFGCN innovatively designs a learnable EEG frequency band tuned module and a heuristic space topology module. These two modules allow for the intricate extraction of both frequency and spatial topological features, utilizing graph adjacency matrices that encapsulate channel correlations and spatial relationships, hence fostering individualized analysis and enhanced generalizability across subjects. Furthermore, a spatio-temporal convolution module paired with a feature map attention mechanism is proposed to extract the critical spatio-temporal features of electroencephalogram (EEG) data. Validation experiments on the PhysioNet and LLM-BCImotion datasets against six mainstream methods demonstrate that THFGCN outperforms state-of-theart methods, achieving 88.41% and 82.82% accuracy in the within-subject case, and 65.93% and 60.56% accuracy in the cross-subject case, respectively. Detailed frequency band weight and T-distributed Stochastic Neighbor Embedding visualization validate the effectiveness of proposed modules. Furthermore, feature interpretability analysis proves the extracted features’ profound MI task relevance, underlining THFGCN’s exceptional interpretability.

12 July 2022 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
FMFengjun Mu
Fengjun Mu
RHRui Huang
Rui Huang
KSKecheng Shi
Kecheng Shi
XLXin Li
Xin Li
JQJing Qiu
Jing Qiu
HCHong Cheng
Hong Cheng
6D object pose estimation is an essential task in vision-based robotic grasping and manipulation. Prior works always train models with a large number of pose annotated images, limiting the efficiency of model transfer between different scenarios. This letter presents an end-to-end model named Weak6D , which could be learned with unannotated RGB-D data. The core of the proposed approach is the novel optimizing method Iterative Annotation Resolver, which has the ability to directly utilize the captured RGB-D data through the training process. Furthermore, we employ a weak refinement loss to optimize the pose estimation network with refined object poses. We evaluated the proposed Weak6D in the YCB-Video dataset, and experimental results show our model achieved practical results without annotated data.

10 September 2024 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
FMFengjun Mu
Fengjun Mu
RHRui Huang
Rui Huang
JZJingting Zhang
Jingting Zhang
CZChao-bin Zou
Chao-bin Zou
KSKecheng Shi
Kecheng Shi
HCHong Cheng
Hong Cheng
Object pose estimation has extensive applications in various industrial scenarios. However, the heavy reliance on dense 6-D annotation and textured object models has become a significant obstacle to the widespread industrial application of 6-D object pose estimation methods. In this work, we present SS-Pose , a self-supervised learning framework for estimating 6-D object poses without annotated 6-D data and textured model. SS-Pose proposes the coordinate system datum reinitializer stage to dynamically establish a sequence-level pose representation datum, and the temporal–spatial constraint resolver module to obtain the self-supervised learning target through interframe constraints. We introduce a one-shot cross-coordinate transformation that establishes the relationship between the 6-D representation and the object poses, which can be further utilized in real-world tasks. We evaluated the proposed SS-Pose on the challenging YCB-Video dataset and texture-less T-LESS dataset. Our approach achieves competitive performance with significantly lower data dependency, making it suitable for visual perception in industrial applications.

24 January 2024 in International Journal of Humanoid Robotics
Verified
Journal Article
Closed access
CZChao-bin Zou
Chao-bin Zou
RHRui Huang
Rui Huang
GSGuangkui Song
Guangkui Song
KSKecheng Shi
Kecheng Shi
Zhinan PengZhinan Peng
Zhinan Peng
HCHong Cheng
Hong Cheng

27 November 2025 in IEEE Access
Verified
Journal Article
Open access
HLHang Li
Hang Li
KSKecheng Shi
Kecheng Shi
FMFengjun Mu
Fengjun Mu
HYHuan Yang
Huan Yang
HLHongyan Lai
Hongyan Lai
In existing EEG signal processing applications, it is generally assumed that different brain regions exhibit consistent connectivity patterns under identical scenarios, with variations attributable to individual or environmental differences. However, recent advances in neuroscience have revealed that functional connectivity (FC) within the brain is inherently dynamic and varies continuously, even for the same individual under identical or continuous scenarios. Current studies often represent inter-regional FC using correlation matrices derived from EEG channels; however, these are typically precomputed as static matrices or iteratively updated under static assumptions. While emerging medical research leverages dynamic FC from fMRI for brain disorder classification, fMRI’s low temporal resolution of fMRI limits its applicability in real-time tasks, such as driver fatigue detection or motor intent recognition. This study constructs dynamic FC features using Pearson correlation coefficients from EEG signals and validates their efficacy in emotion recognition and driver fatigue scenarios, demonstrating significant performance improvements, although good results were obtained in the application of time; no significant differences were directly found in the direct comparison with static brain connections. Therefore, we believe that dynamic brain connections can only be used as prior knowledge of graphs and not as a separate feature to perform classification or recognition tasks.

27 September 2021
Verified
Conference Paper
Closed access
RHRui Huang
Rui Huang
Zhinan PengZhinan Peng
Zhinan Peng
SGSiying Guo
Siying Guo
KSKecheng Shi
Kecheng Shi
CZChao-bin Zou
Chao-bin Zou
HCHong Cheng
Hong Cheng
Estimating the center of mass (CoM) is essential for both gait planning and controlling of lower limb exoskeletons. Different from CoM estimation in human and humanoid robots, a critical issue in human-exoskeleton systems pis how to describe the effect of physical human-exoskeleton interactions in estimating the CoM of lower limb exoskeletons. This paper presents a novel center of mass estimation method Physically Coupled Serial Chain (PCSC) for human-coupled lower limb exoskeleton systems. Different from traditional serial chain methods, the proposed PCSC involves physical human-exoskeleton models to describe physical interactions between the pilot and the lower limb exoskeleton. We demonstrated the effectiveness of proposed PCSC model in the AIDER lower limb exoskeleton system. Experimental results indicate that the proposed PCSC model is more accuracy than traditional serial chain methods.