Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Suggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works180 from public data
- Relational Knowledge Distillation2,135
RKD allows students to outperform their teachers' performance, achieving the state of the arts on standard benchmark datasets and proposes distance-wise and angle-wise distillation losses that penalize structural differences in relations.
- Neural Video Compression with Diverse Contexts363
This paper introduces a group-based offset diversity where the cross-group interaction is proposed for better context mining and adopts a quadtree-based partition to increase spatial context diversity when encoding the latent representation in parallel.
- MonoGRNet: A Geometric Reasoning Network for Monocular 3D Object Localization287
This work proposes a novel IDE method that directly predicts the depth of the targeting 3D bounding box's center using sparse supervision, and demonstrates that MonoGRNet achieves state-of-the-art performance on challenging datasets.
- StableVideo: Text-driven Consistency-aware Diffusion Video Editing243
This paper develops a novel inter-frame propagation mechanism for diffusion video editing, which leverages the concept of layered representations to propagate the appearance information from one frame to the next.
- Virtualized Screen: A Third Element for Cloud–Mobile Convergence170
Two novel cloud-mobile applications, Cloud Browser and Cloud Phone, are presented to demonstrate the advantages of a virtualized screen, and technical challenges are discussed and addressed.
- WBC-Net: A white blood cell segmentation network based on UNet++ and ResNet151
A deep learning network called WBC-Net, which is based on UNet++ and ResNet, is proposed to improve WBC image segmentation accuracy and achieves better WBC segmentation performance than several state-of-the-art methods.
- Rethinking Minimal Sufficient Representation in Contrastive Learning114
This work proposes to increase the mutual information between the representation and input as regularization to approximately introduce more task-relevant information, since the classic contrastive learning models have the risk of overfitting to the shared information between views.
- Affinity Derivation and Graph Merge for Instance Segmentation113
This work presents an instance segmentation scheme based on pixel affinity information, which is the relationship of two pixels belonging to the same instance, which uses two neural networks with similar structures to cluster pixels into instances.
- 108
- Distributed multi-view video coding92
A practical framework of distributed multi-view video coding, in which inter-camera communication is avoided and the large computing complexity is moved from encoder to decoder, and the traditional inter frame is replaced by Wyner-Ziv frame.
- 76
- Reliable Propagation-Correction Modulation for Video Object Segmentation76
This work aims to suppress error propagation through a correction mechanism with high reliability through a cascaded propagation-correction scheme and augments the reference cues by supplementing reliable feature patches to a maintained pool, thus offering more comprehensive and expressive object representations to the modulators.
- In Defense of the Classification Loss for Person Re-Identification75
This paper proposes a person re-id framework featured by channel grouping and multi-branch strategy, which divides global features into multiple channel groups and learns the discriminative channel group features by multi- Branches classification layers.
- Robust Referring Video Object Segmentation with Cyclic Structural Consensus74
This work jointly modeling the primary R-VOS problem and its dual (text reconstruction) and introduces a structural text-to-text cycle constraint to discriminate semantic consensus between video-text pairs and impose it in positive pairs, thereby achieving multi-modal alignment from both positive and negative pairs.
- 74
- 72
- 70
- 64
- 60
- 55
- 53
- 52
- 50
- MVPNet: Multi-View Point Regression Networks for 3D Object Reconstruction from A Single Image49
This paper proposes an encoder-decoder network that generates such kind of multiple view-dependent point clouds from a single image by regressing their 3D coordinates and visibilities and introduces a novel geometric loss that is able to interpret discrepancy over 3D surfaces as opposed to 2D projective planes.
- MMedAgent-RL: Optimizing Multi-Agent Collaboration for Multimodal Medical Reasoning48
This work proposes MMedAgent-RL, a reinforcement learning (RL)-based multi-agent framework that enables dynamic, optimized collaboration among medical agents, and introduces a curriculum learning (CL)-guided RL strategy with dynamic entropy regulation to address the inconsistency in specialist outputs.
- 48
- 47
- Self-Supervised Video Representation Learning with Meta-Contrastive Network47
A Meta-Contrastive Network (MCN) is proposed, which combines the contrastive learning and meta learning, to enhance the learning ability of existing self-supervised approaches.
- 43
- End-to-End Neural Speech Coding for Real-Time Communications42
Both subjective and objective results demonstrate the efficiency of the proposed TFNet, an end-to-end neural speech codec with low latency for RTC that takes an encoder-temporal filtering-decoder paradigm that has seldom been investigated in audio coding.
- Self-adaptive model-based ECG denoising using features extracted by mean shift algorithm41
This paper proposes a mean shift based initializer to provide a much more robust estimation of initial model parameters for each heart beat and demonstrates satisfactory performance for both normal and abnormal heart beats under both white and pink noises.
- Generative Latent Coding for Ultra-Low Bitrate Image and Video Compression40
This work proposes Generative Latent Coding models for image and video compression, termed GLC-image and GLC-Video, and improves the hyper prior by introducing a spatial categorical hyper module in GLC-image and a spatio-temporal categorical hyper module in GLC-video to enhance the semantic consistency.
- Efficient background video coding with static sprite generation and arbitrary-shape spatial prediction techniques39
An efficient static sprite-generation and compression scheme is proposed for background video coding and a reliability-based blending technique is developed in the sprite generation to prevent the sprite image from being blurred due to the inaccurate segmentation.
- 35
- A global survey on the use of the international classification of diseases codes for metabolic dysfunction-associated fatty liver disease34
This global survey by a collaborative panel of clinical, coding, health management and policy experts indicates agreement that MAFLD should be coded in ICD-11.
- Leukocyte Image Segmentation Based on Adaptive Histogram Thresholding and Contour Detection33
Experimental results on two datasets containing 160 leukocyte images show that the proposed method obtains more accurate segmentation results than their counterparts under rapid and standard staining conditions.
- SSAN: Separable Self-Attention Network for Video Representation Learning31
A separable self-attention (SSA) module is proposed, which models spatial and temporal correlations sequentially, so that spatial contexts can be efficiently used in temporal modeling.
- 31
- 30
- Towards Robust Video Object Segmentation with Adaptive Object Calibration29
A new deep network is proposed, which can adaptively construct object representations and calibrate object masks to achieve stronger robustness and achieves the state-of-the-art performance among existing published works, and also exhibits superior robustness against perturbations.
- 29
- 28
- Latent-Domain Predictive Neural Speech Coding28
This article introduces latent-domain predictive coding into the VQ-VAE framework to fully remove temporal redundancies within encoded features and proposes the TF-Codec for low-latency neural speech coding in an end-to-end manner.
- 27
- UniGraspTransformer: Simplified Policy Distillation for Scalable Dexterous Robotic Grasping26
UniGraspTransformer is introduced, a universal Transformer-based network for dexterous robotic grasping that simplifies training while enhancing scalability and performance, and generates a broader range of grasping poses for objects in various shapes and orientations, resulting in more diverse grasp strategies.
- Disentangled Feature Learning for Real-Time Neural Speech Coding26
It is found that the learned disentangled features show comparable performance on any-to-any voice conversion with modern self-supervised speech representation learning models with far less parameters and low latency, showing the potential of the neural coding framework.
- Introduction to the Special Section on Visual Computing in the Cloud: Cloud Gaming and Virtualization23
Cloud gaming, the newest entry in the online gaming world, leverages the well-known concept of cloud computing to provide real-time gaming services to players, with recent successes such as its Kahawai project.
- Real-time video coding under power constraint based on H.264 codec23
A joint power-distortion optimization scheme for real-time H.264 video encoding under the power constraint is translated to the complexity constraint based on DVS technology and a computation allocation model with virtual buffers is proposed to facilitate the optimal allocation of constrained computational resource for each frame.
- 23
- 22
- Hybrid Instance-Aware Temporal Fusion for Online Video Instance Segmentation22
An online video instance segmentation framework with a novel instance-aware temporal fusion method that is able to directly retrieve and maintain instance identities across frames, eliminating the complicated frame-wise instance matching in prior methods.
- 21
- 21
- Video Instance Segmentation by Instance Flow Assembly20
This work proposes a bottom-up framework equipped with a temporal context fusion module to better encode inter-frame correlations and introduces an instance-level correspondence across adjacent frames, represented by a center-to-center flow, termed as instance flow, to assemble messy dense temporal correspondences.
- 20
- 20
- 20
- A Universal Optimization Framework for Learning-based Image Codec19
A plug-and-play optimization framework for seeking higher compression ratio, which can be flexibly applied to existing and potential future compression networks, and a novel latent optimization algorithm to adaptively remove the redundancy for each image is proposed.
- 19
- 19
- Directional Lifting-Based Wavelet Transform for Multiple Description Image Coding with Quincunx Segmentation19
A new multiple description image coding scheme using directional lifting transform to extensively exploit the correlations among neighboring pixels to improve the visual quality from the central decoder.
- Deep learning for tracing esophageal motility function over time18
This study effectively trace esophageal motility function with HRM by using a deep learning computational model, namely, EMD-DL, which leverages three-dimensional convolution (Conv3D) and bidirectional convolutional long-short-term-memory (BiConvLSTM) models.
- A proxy-based mobile web browser18
This paper uses the server-side web parsing and rendering to leverage the browser computing logic and uses a composite screen format to represent the display of the web content, incorporating the web background screen and the dynamic web objects.
- Uncertainty-Aware Deep Video Compression With Ensembles17
This work proposes an uncertainty-aware video compression model that can effectively capture the predictive uncertainty with deep ensembles and introduces an ensemble-aware loss to encourage the diversity among ensemble members and investigate the benefits of incorporating adversarial training in the video compression task.
- Dasformer: Deep Alternating Spectrogram Transformer For Multi/Single-Channel Speech Separation17
Experiments show that DasFormer has a powerful ability to model the time-frequency representation, whose performance far exceeds the current SOTA models in multi-channel speech separation, and also achieves single-channel SOTA in the more challenging yet realistic reverberation scenario.
- 16
- A Study on the Quantization Scheme in H.264/AVC and Its Application to Rate Control16
An efficient rate control scheme is developed based on the new rate-distortion (R-D) model and implemented into the H.264/AVC reference software, with which the better coding performance can be achieved.
- Fast and Robust Sprite Generation for MPEG-4 Video Coding16
The proposed algorithm can significantly improve the visual quality of the generated sprite image and the reconstructed background object in video coding and can be used for both frame-based and object-based video coding.
- 15
- 15
- Free viewpoint switching in multi-view video streaming using Wyner-Ziv video coding15
A novel free viewpoint switching scheme for multi-view video scenario, in which the distributed video coding technique is employed, and the inter-camera communication can be avoided and the drifting error can be controlled efficiently when the viewpoint switching occurs.
- 15
- 14
- 14
- 14
- 14
- Residual Refinement Network with Attribute Guidance for Precise Saliency Detection13
This work strengthens a residual refinement network with feature purification for better saliency estimation and introduces an attribute encoding sub-network (AENet) with a grid aggregation block (GAB) to guide the final saliency predictor to obtain more accurate saliency maps.
- B-picture coding in AVS video compression standard13
The proposed symmetrical mode that replaces the conventional bi-directional mode, in which only one motion vector is coded and another is derived from the coded one with the assumption of approximate constant-speed motion and improves the accuracy of derived motion vectors in temporal direct mode with division-free operations.
- Recursive Least-Squares Estimator-Aided Online Learning for Visual Tracking12
This paper proposes a simple yet effective recursive least-squares estimator-aided online learning approach for few-shot online adaptation without requiring offline training that allows an in-built memory retention mechanism for the model to remember the knowledge about the object seen before, and thus the seen data can be safely removed from training.
- 12
- 11
- 11
- A High-Fidelity and Low-Interaction-Delay Screen Sharing System11
A layered structure for screen coding and rendering is proposed to deliver diverse screen content to the client side with an adaptive frame rate to enable screen sharing among multiple devices with high fidelity and responsive interaction.
- 10
- 9
- 9
- 9
- 9
- 8
- 8
- 8
- 8
- 8
- 8
- ForenX: Towards Explainable AI-Generated Image Detection with Multimodal Large Language Models7
ForenX is presented, a novel method that not only identifies the authenticity of images but also provides explanations that resonate with human thoughts that enhance the generalization of forgery detection and provide explanations that are accurate, relevant, and comprehensive.
- 7
- 7
- 7
- 7
- 7
- Depth sensor assisted real-time gesture recognition for interactive presentation7
A gesture recognition approach to enable real-time manipulating projection content through detecting and recognizing speakers gestures from the depth maps captured by a depth sensor is presented.
- Towards Anytime Retrieval: A Benchmark for Anytime Person Re-Identification6
A unified model named Uni-AT is proposed, which comprises a multi-scenario ReID (MS-ReID) framework for scenario-specific features learning, a Mixture-of-Attribute-Experts (MoAE) module to alleviate inter-scenario interference, and a Hierarchical Dynamic Weighting (HDW) strategy to ensure balanced training across all scenarios.
- 6
- 6
- 6
- Phoneme-Based Distribution Regularization for Speech Enhancement6
A phoneme-based distribution regularization (PbDr) for speech enhancement, which incorporates frame-wise phoneme information into speech enhancement network in a conditional manner, and can not only boost the perceptual quality forspeech enhancement but also the recognition accuracy of an ASR system on the enhanced speech.
- 6
- 6
- 6
- 5
- 5
- 5
- ReDi5
Evaluations show that the proposed system can efficiently utilize the remote computing resources and local display capabilities of ubiquitous devices, which will greatly benefit interactive multimedia applications.
- Rate-distortion optimized color quantization for compound image compression5
A rate-distortion optimized color quantization algorithm for text/graphic content, which introduces distortion to text content and minimizes the bit rate produced by the following lossless entropy compression algorithm.
- 5
- MLLM-based Discovery of Intrinsic Coordinates and Governing Equations from High-Dimensional Data4
This paper designs a series of enhanced visual prompts for MLLM to enhance its spatial perception, and proposes a zero-shot method based on MLLM for automatically discovering physical coordinates and governing equations from high-dimensional data.
- 4
- 4
- 4
- 4
- Embedded image coding based on novel organization of DCT coefficients4
In addition to the well known property of energy compaction, novel organization of block-based DCT coefficients has similar characteristics to wavelet transform when taking an image as a single DCT clustering entity and can be exploited to obtain higher compression performance and to widen DCT applications.
- Deciphering Functions of Neurons in Vision-Language Models3
It is found that there are neurons responsible for only visual or text information, or both, respectively, which are referred to them as visual neurons, text neurons, and multi-modal neurons, respectively.
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 2
- Intelligent Prediction of Network Security Situations based on Deep Reinforcement Learning Algorithm2
- 2
- Fast Video Stitching for Aerially Captured HD Videos2
This work proposes a real-time video stitching system which is capable to stitch high definition (HD) videos captured by mobile aerial devices and uses scale invariant information to speed up the feature point extraction.
- 2
- 2
- 2
- 2
- 2
- 2
- Design of a wavelet slave processor for audio and video decompression2
The intention of this paper is to present an architecture for a WT slave processor, an architectural design for an integer Wavelet Slave Processor (WSP) for decompression and to show the effectiveness of wavelet transform.
- 1
- Text-Queried Audio Source Separation via Hierarchical Modeling1
A hierarchical decomposition framework, HSM-TSS, that decouples the task into global-local semantic-guided feature separation and structure-preserving acoustic reconstruction, achieving state-of-the-art separation performance with data-efficient training while maintaining superior semantic consistency with queries in complex auditory scenes.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Content adaptive screen image scaling1
This paper proposes an efficient content adaptive screen image scaling scheme for the real-time screen applications like remote desktop and screen sharing that can achieve good visual quality and also keep the low complexity for realtime applications.
- 1
- 1
- 1
- Scalable Wyner-Ziv video coding with adaptive bit-plane representation1
A simplified adaptive bit-plane representation is proposed without pre-knowing the Laplacian distribution parameters for the DCT-domain Wyner-Ziv video coding, in which the encoding can stop at any bit-planes and the bit-stream can also be flexibly truncated.
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Distributed Video Coding with Trellis Coded Quantization–
Experimental results indicate that though the simplest TCQ is used, gain is also achieved over scalar quantization, and a method to solve the refinement problem is proposed.
- –
- –
- –
- –
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 2026-10-16. 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.