Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks95 from public data
- 223
- Hierarchical Dense Correlation Distillation for Few-Shot Segmentation186
This work designs Hierarchically Decoupled Matching Network (HDMNet) mining pixel-level support correlation based on the transformer architecture and proposes a matching module to reduce train-set overfitting and introduce correlation distillation leveraging semantic correspondence from coarse resolution to boost fine-grained segmentation.
- 97
- 79
- Temporal Enhanced Training of Multi-view 3D Object Detector via Historical Object Prediction77
This paper proposes a new paradigm, named Historical Object Prediction (HoP), for multi-view 3D detection to leverage temporal information more effectively and can be easily incorporated into state-of-the-art BEV detection frameworks, including BEVFormer and BEVDet series.
- A Deep Transfer Learning Network for Structural Condition Identification with Limited Real-World Training Data77
The results show that even though the training data are from a different domain and with different types of labels, intrinsic physics can be learned through the pretraining process, and the TL results can be clearly improved, with the identification accuracy increasing from 81.8% to 89.1%.
- A Labeling Method for Financial Time Series Prediction Based on Trends77
Experiments showed that the proposed continuous trend labeling method is a much better state-of-the-art labeling method in terms of classification accuracy and some other classification evaluation metrics, and proved that deep learning models such as LSTM and GRU are more suitable for dealing with the prediction of financial time series data.
- Fitting smoothing splines to time-indexed, noisy points on nonlinear manifolds60
This paper derives certain geometrical elements, namely the exponential map and its inverse, parallel transport of tangents, and the curvature tensor, on these manifolds that are needed in the gradient-based search for smoothing splines.
- Elastic Functional Coding of Riemannian Trajectories59
The TSRVF representation, and accompanying statistical summaries of Riemannian trajectories, are utilized to extend existing coding methods such as PCA, KSVD and Label Consistent KSVD to Riemanni trajectories or more generally to R Siemannian functions, showing that such coding efficiently captures trajectories in applications such as action recognition, stroke rehabilitation, visual speech recognition, clustering and diverse sequence sampling.
- Vision-Language Gradient Descent-driven All-in-One Deep Unfolding Networks46
VLU-Net is the first all-in-one DUN framework and outperforms current leading one-by-one and all-in-one end- to-end methods by 3.74 dB on the SOTS dehazing dataset and 1.70 dB on the Rain100L deraining dataset.
- 46
- 40
- Semi-supervised medical image segmentation via weak-to-strong perturbation consistency and edge-aware contrastive representation38
This paper proposes an innovative framework that effectively leverages unlabeled data to improve segmentation performance, especially in edge regions and introduces a weak-to-strong perturbation strategy with corresponding feature-perturbed consistency loss to efficiently utilize unlabeled data.
- Two-Person Graph Convolutional Network for Skeleton-Based Human Interaction Recognition38
A novel unified two-person graph to represent inter-body and intra-body correlations between joints and several graph labeling strategies are designed to supervise the model to learn discriminant spatial-temporal interactive features.
- 36
- Statistical analysis of manual segmentations of structures in medical images34
This paper uses statistical shape theory to generate joint inferences and analyze data generated by a new paradigm in medical image segmentation, termed Citizen Science, which involves a volunteer effort from multiple, possibly non-expert, human participants to estimate anatomical boundaries in the form of planar closed curves.
- 32
- A BERT Based Method for Continuous Estimation of Cross-Subject Hand Kinematics From Surface Electromyographic Signals32
The proposed method based on Bidirectional Encoder Representation from Transformers (BERT) structure to predict the movement of hands from the root mean square (RMS) feature of the sEMG signal following $\mu $ -law normalization achieved state-of-the-art performance in cross-subject evaluation as well as high performance in subject-specific tests on the Ninapro dataset.
- 31
- 31
- Rate-Invariant Analysis of Covariance Trajectories29
This work represents covariance trajectories using transported square-root vector fields, constructed by parallel translating scaled-velocity vectors of trajectories to their starting points, using a natural Riemannian metric on this vector bundle, which leads to a rate-invariant analysis of trajectory analysis.
- Online and offline based load balance algorithm in cloud computing27
A new load balancing method combined with the advantage of online and offline load balancing algorithms are proposed in this paper and good performance shows the efficiency of the proposed method.
- 24
- 24
- Detection, classification and estimation of individual shapes in 2D and 3D point clouds24
A generalized likelihood ratio test for detecting a shape in a point cloud, based on optimizing over some unknown parameters, including the pose and scale associated with hypothesized objects, and an empirical evaluation of the log-likelihood ratio distribution is developed.
Publication data from OpenAlex; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.