Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks65 from public data
- Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers3,852
This paper deploys a pure transformer to encode an image as a sequence of patches, termed SEgmentation TRansformer (SETR), and shows that SETR achieves new state of the art on ADE20K, Pascal Context, and competitive results on Cityscapes.
- Long-Term Cloth-Changing Person Re-identification248
This work introduces a shape embedding module as well as a cloth-elimination shape-distillation module aiming to eliminate the now unreliable clothing appearance features and focus on the body shape information.
- Simpler is Better: Few-shot Semantic Segmentation with Classifier Weight Transformer243
This work proposes to simplify the meta-learning task by focusing solely on the simplest component – the classifier, whilst leaving the en-coder and decoder to pre-training, and introduces a Classifier Weight Transformer (CWT) designed to dynamically adapt the support-set trained classifier’s weights to each query image in an inductive way.
- Learning a Few-shot Embedding Model with Contrastive Learning207
This paper proposes a novel contrastive training scheme dubbed infoPatch, exploiting the patch-wise relationship to substantially improve the popular infoNCE and investigates the contrastive learning with Noise Contrastive Estimation in a supervised manner for training a few-shot embedding model.
- Instance Credibility Inference for Few-Shot Learning199
This paper presents a simple statistical approach, dubbed Instance Credibility Inference (ICI), to exploit the distribution support of unlabeled instances for few-shot learning to establish new state-of-the-arts on four widely used few- shot learning benchmark datasets.
- Depth-conditioned Dynamic Message Propagation for Monocular 3D Object Detection170
A depth-conditioned dynamic message propagation (DDMP) network to effectively integrate the multi-scale depth information with the image context and dynamically predicting hybrid depth-dependent filter weights and affinity matrices for propagating information is proposed.
- Learning Dynamic Alignment via Meta-filter for Few-shot Learning134
This paper proposes to learn a dynamic alignment, which can effectively highlight both query regions and channels according to different local support information and facilitates dynamical knowledge adaptation for few-shot learning.
- SeaFormer++: Squeeze-Enhanced Axial Transformer for Mobile Visual Recognition132
A new method squeeze-enhanced Axial Transformer (SeaFormer) for mobile visual recognition, designed a generic attention block characterized by the formulation of squeeze Axial and detail enhancement, which can be further used to create a family of backbone architectures with superior cost-effectiveness.
- Depth Guided Adaptive Meta-Fusion Network for Few-shot Video Recognition124
A depth guided Adaptive Meta-Fusion Network for few-shot video recognition which is termed as AMeFu-Net, which is trained in the meta-learning way and demonstrates the effectiveness of the model on several action recognition benchmarks.
- 114
- Effects of various LED light wavelengths and intensities on the performance of purifying synthetic domestic sewage by microalgae at different influent C/N ratios112
It is demonstrated that the microalgae Chlorella vulgaris can effectively remove nutrients from domestic sewage using the optimal influent C/N ratio, which is 5:1.
- ONCE-3DLanes: Building Monocular 3D Lane Detection98
This work presents ONCE-3DLanes, a real-world autonomous driving dataset with lane layout annotation in 3D space, and presents an extrinsic-free, anchorfree method, called SALAD, regressing the 3D coordinates of lanes in image view without converting the feature map into the bird's-eye view (BEV).
- 94
- 92
- Generative Semantic Segmentation77
Generative Semantic Segmentation (GSS), a generative learning approach for semantic segmentation, is presented by replacing the conventional per-pixel discriminative learning with a latent prior learning process that model the variational posterior distribution of latent vari-ables given the segmentation mask.
- RoDyn-SLAM: Robust Dynamic Dense RGB-D SLAM With Neural Radiance Fields74
This work proposes a novel dynamic SLAM framework with neural radiance field that achieves state-of-the-art performance among recent neural RGB-D methods in both accuracy and robustness and introduces a motion mask generation method to filter out the invalid sampled rays.
- Delving into Data: Effectively Substitute Training for Black-box Attack73
A novel perspective substitute training that focuses on designing the distribution of data used in the knowledge stealing process and the combination of these two modules can further boost the consistency of the substitute model and target model, which greatly improves the effectiveness of adversarial attack.
- Boundary-sensitive Pre-training for Temporal Localization in Videos66
This paper proposes to synthesize temporal boundaries in existing video action classification datasets and introduces a novel boundary-sensitive pretext (BSP) task, which is superior and complementary to the existing action classification-based pre-training counterpart and achieves new state-of-the-art performance on several temporal localization tasks.
- 57
- How to Trust Unlabeled Data? Instance Credibility Inference for Few-Shot Learning55
A statistical approach, dubbed Instance Credibility Inference (ICI), to exploit the support of unlabeled instances for few-shot visual recognition to collect all the correctly-predicted pseudo-labeled instances from the noisy pseudo- labeled set.
- Imaging port wine stains by fiber optical coherence tomography51
It is demonstrated that OCT may be a useful tool for the noninvasive imaging of PWS and may help determine the photosensitizer dose and laser parameters in photodynamic therapy for treating port wine stains.
- S-NeRF++: Autonomous Driving Simulation via Neural Reconstruction and Generation46
S-NeRF++ is an enhanced neural radiance field for synthesizing large-scale scenes and moving vehicles, with improved scene parameterization and camera pose learning, and an advanced foreground-background fusion pipeline that skillfully integrates illumination and shadow effects, further enhancing the realism of the simulations.
- SGM3D: Stereo Guided Monocular 3D Object Detection42
This paper proposes a stereo-guided monocular 3D object detection framework, dubbed SGM3D, which aligns the monocular feature to the stereo feature and exploits the network's ability of generating the cross-modal feature for accurate 3Dobject detection.
- A Unified Efficient Pyramid Transformer for Semantic Segmentation35
This work proposes a unified framework (UN-EPT) to segment objects by considering both context information and boundary artifacts, and demonstrates promising performance on three popular benchmarks for semantic segmentation with low memory footprint.
- 34
- 34
- Vision Transformers: From Semantic Segmentation to Dense Prediction29
This work explores the global context learning potentials of ViTs for dense visual prediction and forms a family of Hierarchical Local-Global (HLG) Transformers, characterized by local attention within windows and global-attention across windows in a pyramidal architecture.
- Softmax-Free Linear Transformers26
A family of SOftmax-Free Transformers (SOFT) is proposed, with a Gaussian kernel function adopted to replace the dot-product similarity, enabling a full self-attention matrix to be approximated under low-rank matrix decomposition.
- 17
- 17
- Dynamic Graph Message Passing Networks for Visual Recognition16
This paper proposes a dynamic graph message passing network, that significantly reduces the computational complexity compared to related works modelling a fully-connected graph, and outperforms fully- connected graphs while using substantially fewer floating-point operations and parameters.
- 13
- 12
- LaneCorrect: Self-Supervised Lane Detection10
This paper proposes a novel self-supervised training scheme that automatically corrects the lane label by learning geometric consistency and instance awareness from the adversarial augmentations, and presents a novel self-supervised pre-trained model.
- 10
- Dynamic Depth Fusion and Transformation for Monocular 3D Object Detection10
This work designs an Adaptive Depth-guided Instance Normalization layer to leverage depth features to guide RGB features for high quality estimation of 3D properties and introduces a Dynamic Depth Transformation module to better recover accurate depth according to semantic context learning.
- Preconditioned Score-Based Generative Models8
A novel preconditioned diffusion sampling (PDS) method that leverages matrix preconditioning to alleviate the aforementioned problem of ill-conditioned issues of the Langevin dynamics and reverse diffusion in the sampling process.
- ImpDet: Exploring Implicit Fields for 3D Object Detection8
This work introduces a new perspective that views bounding box regression as an implicit function, which leverages implicit field learning for 3D object detection and presents a simple yet efficient virtual sampling strategy to solve the problem of sparsity on the object surface.
- Application of Artificial Intelligence Technology in College Physical Education6
The teaching methods of artificial intelligence technology such as virtual reality technology, big data analysis technology and intelligent recognition technology are put forward as the innovation points, in order to solve the shortcomings of traditional sports teaching method.
- 5
- 4
- 3
- 3
- Few-shot Learning for Multi-Modality Tasks3
To simulate the learning schema of humans and relieve the reliance on the large-scale annotation benchmarks, researchers start shifting towards the few-shot learning problem: they try to learn a model to correctly recognize novel categories with only a few annotated samples.
- 3
- PARTNER: Level Up the Polar Representation for 3D Object Detection2
Partner alleviates the dilemma of feature distortion with global representation re-alignment and facilitates the regression by introducing instance-level geometric information into the detection head, enabling a unified framework for multi-modal detection in polar space.
- 2
- 2
- 1
- Exploration on Context-Sensitive Affect Sensing in an Intelligent Agent1
This work equips the AI agent with the capabilities of the inferences of the improvisational mood of individual characters and emotions embedded in communication context to further verify the affect drawn solely from the analysis of each turn-taking input itself.
- 1
- Emotion classification of teaching resources in online teaching based on artificial intelligence–
The experimental results show that the W-LDA model outperforms traditional LDA models in terms of predictive accuracy for topic distribution, and the sentiment analysis method based on topic distribution effectively uncovers latent information within course review texts, providing valuable insights for improving the quality of online teaching.
- MS-Road: Towards Spatiotemporal-Consistent Large-Scale Road Reconstruction–
MS-Road is proposed, a camera-based approach for large-scale road surface reconstruction with strong geometric consistency leveraging self-supervised learning and multi-view consistent constrain to tackle two key issues: inconsistency in road appearance across different observations and inaccurate road height localization.
- –
- –
- Exploring the relationship between functional structure and ecosystem multifunctionality requires intraspecific trait variability–
It is found that the effect of trait variability on multifunctionality depended not only on the number of functions, but also on the thresholds considered, which meant that the relative contribution of intraspecific variability would be more important.
- –
- Comparison of perioperative outcomes in robot-assisted partial nephrectomy and laparoscopic partial nephrectomy–
Early comparative outcomes suggest that RAPN has a significant benefit over the LPN in terms of OT,WIT and hospital stay and offers equivalent prognosis and postoperative renal function preservation compared with LPN.
- –
- Semantic-inspired Contextual Affect Detection from Drama Improvisation–
This paper discusses how context is used to interpret affect implied in conversational inputs with weak or no affect indicators embedded in multithreaded social interactions.
- –
- Studying tumor metastasis by in vivo imaging and flow cytometer–
The " in vivo microscopy" is developed to study the mechanisms that govern liver tumor cell spread through the microenvironment in vivo in real-time confocal near-infrared fluorescence imaging and a recently developed "in vivo flow cytometer" and optical imaging are used to assess Liver tumor cell spreading and the circulation kinetics of liver tumor cells.
- Well Separated Pair Decomposition–
This lecture introduces the notion of well-separated pair decomposition (WSPD) that clusters the set of all pairwise distances in a given point set and discusses a few applications of WSPD.
- –
- An Improvisational AI Actor in E-Drama.–
A method of target acquisition and lock-on-launch strike capability of self guided explosive canisters by launching an imaging sensor platform over the battlefield area and transmitting imagery of the battlefield to an image processing computer system and an image receiving on-board microcomputer.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. 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.