Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks97 from public data
- LL3DA: Visual Interactive Instruction Tuning for Omni-3D Understanding, Reasoning, and Planning290
LL3DA is presented, a Large Language 3D Assistant that takes point cloud as the direct input and responds to both text instructions and visual interactions that enables LMMs to better comprehend human interactions with the 3D environment and further remove the ambiguities within plain texts.
- Vision-model-based impairment metric to evaluate blocking artifacts in digital video136
Investigations are conducted to simplify and refine a vision-model-based video quality metric without compromising its prediction accuracy and the results show a strong correlation between the objective blocking ratings and the mean opinion scores on blocking artifacts.
- CTransCNN: Combining transformer and CNN in multilabel medical image classification120
The framework developed exhibits strong competitiveness compared to previous research and its robust generalization ability makes it applicable to other medical multilabel image classification tasks.
- FECANet: Boosting Few-Shot Semantic Segmentation With Feature-Enhanced Context-Aware Network109
A feature enhancement module is proposed to suppress the matching noise caused by inter-class local similarity and enhance the intra-class relevance in the naive correlation, and a novel correlation reconstruction module is proposed that encodes extra correspondence relations between foreground and background and multi-scale context semantic features.
- Scene text extraction based on edges and support vector regression100
This paper presents a scene text extraction technique that automatically detects and segments texts from scene images by using a support vector regression model that is trained using bags-of-words representation.
- MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer77
A novel framework named Multimodal Alignment-Guided Dynamic Token Pruning (MADTP) for accelerating various VLTs is proposed, which significantly reduces the computational complexity of kinds of multimodal models while preserving competitive performance.
- Saliency Guided Inter- and Intra-Class Relation Constraints for Weakly Supervised Semantic Segmentation71
A saliency guided class-agnostic distance module to pull the intra-category features closer by aligning features to their class prototypes and an object guided label refinement module to take a full use of both the segmentation prediction and the initial labels for obtaining superior pseudo-labels.
- Semantically Meaningful Class Prototype Learning for One-Shot Image Segmentation60
This paper proposes to leverage the multi-class label information during the episodic training to encourage the network to generate more semantically meaningful features for each category and proposes a fused prototype guidance branch for the segmentation of the query image.
- Spatial Structure Constraints for Weakly Supervised Semantic Segmentation51
A CAM-driven reconstruction module to directly reconstruct the input image from deep CAM features, which constrains the diffusion of last-layer object attention by preserving the coarse spatial structure of the image content, and an activation self-modulation module to refine CAMs with finer spatial structure details by enhancing regional consistency are proposed.
- Multi-Granularity Denoising and Bidirectional Alignment for Weakly Supervised Semantic Segmentation50
This work proposes an end-to-end multi-granularity denoising and bidirectional alignment model, to alleviate the noisy label and multi-class generalization issues, and proposes the online noise filtering and progressive noise detection modules to tackle image-level and pixel-level noise.
- IQAGPT: computed tomography image quality assessment with vision-language and ChatGPT models44
This study introduces IQAGPT, an innovative computed tomography IQA system that integrates image-quality captioning VLM with ChatGPT to generate quality scores and textual reports and demonstrates the feasibility of assessing image quality using LLMs.
- 42
- 39
- Towards an end-to-end artificial intelligence driven global weather forecasting system37
It is demonstrated that Adas can assimilate global conventional observations to produce high-quality analysis, enabling the system to operate stably for long term, and can generate accurate end-to-end weather forecasts with comparable skill to those of the IFS, demonstrating the promising potential of data-driven approaches.
- 35
- VideoMAC: Video Masked Autoencoders Meet ConvNets33
This paper proposes a new approach termed as VideoMAC, which combines video masked autoencoders with resource-friendly Con-vNets, and demonstrates that VideoMAC, empowering classical / modern convolutional encoders to harness the benefits of MVM, outperforms ViT-based approaches on downstream tasks.
- 25
- Detection and Grading of Gliomas Using a Novel Two-Phase Machine Learning Method Based on MRI Images24
- 22
- Hyperscale Hardware Optimized Neural Architecture Search22
This paper introduces the first Hyperscale Hardware Optimized Neural Architecture Search (H2O-NAS) to automatically design accurate and performant machine learning models tailored to the underlying hardware architecture to facilitate a smooth and automated transition from research to production.
- A computer-aided diagnosis system for brain tumors based on artificial intelligence algorithms22
A computer-aided diagnosis system for glioma detection, grading, segmentation, and knowledge discovery based on artificial intelligence algorithms is built and offers tumor visualization using a semi-automatic segmentation tool for better patient management and treatment monitoring.
- 21
- 21
- 19
- 19
- 17
- Gene prediction by the noise-assisted MEMD and wavelet transform for identifying the protein coding regions17
A method on basis of the noise-assisted multivariate empirical mode decomposition (NA-MEMD) and the modified Gabor-wavelet transform (MGWT) that can enhance the spectra of the analyzed sequences, and improve the robustness of MGWT to different DNA sequences, thus obtaining higher identification accuracies of protein coding regions over other applied methods.
- 17
- 16
- 15
- 14
- Enhancing Robustness in Learning with Noisy Labels: An Asymmetric Co-Training Approach14
An asymmetric training framework in which one model is robustly trained with a selected subset of clean samples while the other is conventionally trained using the entire training set is introduced, establishing asymmetric sample selection and mining.
- Automated object recognition in high-resolution optical remote sensing imagery13
The background and results of the Automated Object Recognition in Optical Remote Sensing Imagery, which is one of the tracks in 2022 International Algorithm Case Competition, are reported and the challenges, champion solutions, and future directions are summarized.
- A Unified Framework for Biphasic Facial Age Translation With Noisy-Semantic Guided Generative Adversarial Networks11
This work proposes a unified framework for biphasic facial age translation with noisy-semantic guided generative adversarial networks, and introduces two sub-networks, ProjectionNet and ConstraintNet.
- EADNet: Efficient Asymmetric Dilated Network For Semantic Segmentation11
This paper proposes an efficient asymmetric dilated semantic segmentation network, named EADNet, which consists of multiple developed asymmetric convolution branches with different dilation rates to capture the variable shapes and scales information of an image.
- 11
- 9
- 9
- Fast mapping algorithm for histogram to binary set conversion8
An efficient mapping algorithm justifies the practical usage of the prefiltering technique in the application to histogram-based image retrieval systems, especially to searching large image databases.
- Relating CNN-Transformer Fusion Network for Change Detection7
RCTNet introduces an early fusion backbone to exploit both spatial and temporal features early on, a Cross-Stage Aggregation module for enhanced temporal representation, and an Efficient Self-deciphering Attention module utilizing transformers to capture global information and fine-grained details for accurate change detection.
- Semi-Supervised Semantic Segmentation With Region Relevance7
A Region Relevance Network (RRN) is proposed that leverages discriminator networks to assess the accuracy of the pseudo-label at the region level and takes the merit of network diversity to rate the reliability of pseudo-labels and correct the convergence direction of the segmentation network with a dynamic region loss.
- 7
- 7
- BIFRÖST: 3D-Aware Image compositing with Language Instructions6
Bifr\"ost", a novel 3D-aware framework that is built upon diffusion models to perform instruction-based image composition, significantly outperforms existing methods, providing a robust solution for generating realistically composited images in scenarios demanding intricate spatial understanding.
- 6
- 6
- Image Watermarking method using integer-to-integer wavelet transforms6
Simulation results show that the embedded watermark is perceptually invisible and robust to various operations, such as low quality joint picture expert group (JPEG) compression, random and Gaussian noises, and smoothing (mean filtering).
- 6
- One-Shot Learning for Pose-Guided Person Image Synthesis in the Wild5
A Visual Consistency Module (VCM) is introduced, which enhances appearance consistency by combining the face, text, and image embedding in a pre-trained Text2Image model, offering greater stability than state-of-the-art data-driven methods.
- Predicting Spatial Transcriptomics from H&E Image by Pretrained Contrastive Alignment Learning5
CarHE (Contrastive Alignment of gene expRession for hematoxylin and eosin image) is introduced, a method that overcomes limitations by accurately predicting high-dimensional ST data (over 10,000 genes) solely from readily available H&E (Hematoxylin and Eosin) stained images.
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- Hierarchical Graph Pattern Understanding for Zero-Shot VOS4
A new hierarchical graph neural network (GNN) architecture, dubbed hierarchical graph pattern understanding (HGPU), is proposed for zero-shot video object segmentation (ZS-VOS), inspired by the strong ability of GNNs in capturing structural relations.
- 4
- A high-efficient and low-cost secure AMBA framework utilizing configurable data encryption modeling against probe attacks4
Experimental results show that the proposed AMBA framework provides a stronger resistance compared to the other countermeasures and meets the security requirement of total SoC.
- 4
- Generative Image Inpainting with Dilated Deformable Convolution4
A Wasserstein generative adversarial network with dilated convolution and deformable convolution (DDC-WGAN) is proposed for image completion, which enlarges the receptive field and provides a more stable representation of geometric deformation.
- 4
- 4
- 3
- 3
- Application of HDFS cloud storage in digital resource integration of university3
It aims to solve university' problem of massive data storage and construct a cloud storage service system based on HDFS which fit university environment and a digital resource integration method with low storage cost, high efficiency and reliability is studied.
- 2
- Holistic Prototype Attention Network for Few-Shot VOS2
A holistic prototype attention network (HPAN) is proposed for advancing FSVOS, transferring informative knowledge from seen to unseen classes, and exploiting the holistic information from support images and video frames to achieve support-query semantic consistency and inner-frame temporal consistency.
- Application of Random Region Augmentation Algorithm in Deep Learning2
The RRA augmentation algorithm is proposed, which divides four quadrant regions in the image, and randomly selects the ROI region in each region, and is different from the Cutout algorithm directly discarding the region, but randomizing the region.
- Attention-Based DSM Fusion Network for Semantic Segmentation of High-Resolution Remote-Sensing Images2
An attention-based DSM fusion network (ADF-Net) is proposed, which can selectively gather features from spectral and DSM information by channel attention mechanism, and further combine them to get high-quality fusion features.
- Multiple Blind Watermark Algorithm Based on Wavelet Packet Transform and Two-Dimensional Chaos Scramble2
Experimental results show that the multiple blind watermark algorithm based on wavelet packet transform and two-dimensional chaos image scramble has good invisibleness, security and robustness to common image processing and noise attack.
- 2
- Multi-label classification of tongue images using label semantic embedding and dual-branch network1
A Dual-Branch Label Semantic Embedding Network (DBLSEN) model designed to enhance classification accuracy and feature extraction precision in TCM diagnosis, demonstrating the model’s robust generalization capabilities and its potential to significantly advance the field of TCM diagnosis.
- Energy-Efficient Reconfigurable Acceleration Engine for Polynomial Coefficient Generation of Lattice-Based Post-Quantum Cryptography1
This paper investigates and implements an efficient coefficient generation acceleration engine for multiple algorithms based on discrete Gaussian sampling-based and hash function-based coefficient generation schemes separately and proposes an energy-efficient parallel algorithm for the generation of reconfigurable coefficients.
- Learning Physical Dynamics for Object-centric Visual Prediction1
An unsupervised object-centric prediction model that makes future predictions by learning visual dynamics between objects is proposed that generates higher visual quality and more physically reliable predictions compared to the state-of-the-art methods.
- 1
- 1
- –
- Multi-label tongue image classification via label semantics-driven feature interaction enhancement–
A multilabel tongue image classification framework based on Label Semantic-Driven Feature Interaction Enhancement (LSDFIE), which effectively bridges the semantic gap between modalities and leverages hierarchical feature representations comprehensively.
- –
- –
- –
- DynaPlane-Lane: Dynamic Multi-Plane Geometry Learning for Robust Monocular 3D Lane Detection–
DynaPlane-Lane is proposed, a geometry-aware framework that explicitly models road topology through a dynamic multi-plane representation that achieves robust performance across diverse scenarios, validating the effectiveness of incorporating adaptive geometry and structural constraints into monocular 3D lane detection.
- –
- –
- Class-related graph convolution for weakly supervised semantic segmentation–
This work proposes the module of class-related graph convolution, which can perform global modeling and reasoning on the relationship between regions, which is beneficial for many computer vision tasks and yields state-of-the-art performance.
- –
- –
- Fast image annotation based on hash table structure and image segmentation–
The experimental results of both large image dataset NUS-WIDE with 269 648 images and Corel dataset with 5 000 images show that the proposed algorithm can obtain higher annotation speed and annotation accuracy.
- Efficient intra-frame coding algorithm for the depth image in 3D video–
Simulation results show that the proposed method can reduce the coding time by 37.16% on average compared with the independent coding method of depth images, while maintaining high virtual rendering performance.
- Parity quantization of DCT coefficient on text information hiding–
This text proposes parity quantization and amendment algorithm of text information hiding based on DCT domain of image that is robust and the hiding text is invisible, and excellent in resisting the JPEG compression resistance, noise and clipping.
- –
- Study of a New Organizational Method of the Chaos Neural Network–
Usage of image as large-scale chaotic neural network weight matrix was proposed so as to change the organization of neural network, it was found that different images output different curves(surfaces) while identical or similar images output basically the same.
- Design and Implementation of Network Service Performance Monitoring System–
The test results show that the system can meet the demands of monitoring network service performance on the Web-based client and the calculation of the network service response time was realized with the socket connection technique.
- Fast hierarchical motion estimation algorithm based on vector-weighted median filtering–
Simulation results show that the proposed algorithm requires less computational complexity while achieving good performance and a modified summed absolute difference criterion was used for motion estimation of the top level, which could smooth the motion vector field and resulted in an improvement of estimation accuracy.
- Reduction of blocking artifacts by adaptive postfiltering of transform coefficients–
Experimental results show that the proposed technique provides superior performance to other postprocessing methods in both objective and subjective image quality.
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.