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.
Works117 from public data
- 199
- Supervised Raw Video Denoising With a Benchmark Dataset on Dynamic Scenes154
This paper creates motions for controllable objects, such as toys, and captures each static moment for multiple times to generate clean video frames, resulting in the first dynamic video dataset with noisy-clean pairs.
- 135
- 134
- 126
- 123
- 122
- 102
- 101
- rPPG-MAE: Self-Supervised Pretraining With Masked Autoencoders for Remote Physiological Measurements98
A linear self-supervised reconstruction task was designed for extracting the inherent self-similar priors in physiological signals and a specific noise-insensitive strategy was explored for reducing the interference of motion and illumination.
- Virtual-scanning light-field microscopy for robust snapshot high-resolution volumetric imaging94
It is shown that VsLFM achieves ultrafast 3D imaging of diverse processes such as the beating heart in embryonic zebrafish, voltage activity in Drosophila brains and neutrophil migration in the mouse liver at up to 500 volumes per second.
- Learning-based denoising for polarimetric images80
The proposed learning-based method is based on the residual dense network and can significantly suppress the noise in polarimetric images and well reconstruct the details flooded in strong noise.
- 78
- 74
- A Simple Yet Effective Network Based on Vision Transformer for Camouflaged Object and Salient Object Detection72
A simple yet effective network (SENet) based on vision Transformer, employing a simple design of an asymmetric ViT-based encoder-decoder structure that yield competitive results on both tasks, exhibiting greater versatility than meticulously crafted ones is proposed.
- 71
- 59
- NTIRE 2023 Challenge on Light Field Image Super-Resolution: Dataset, Methods and Results52
The first NTIRE challenge on light field (LF) image super-resolution (SR), which aims at super-resolving LF images under the standard bicubic degradation with a magnification factor of 4, develops a new LF dataset called NTIRE-2023 for validation and test, and provides a toolbox called BasicLFSR to facilitate model development.
- 49
- Textured Image Demoiréing via Signal Decomposition and Guided Filtering46
A novel textured image demoiréing method by signal decomposition and guided filtering that outperforms the state-of-the-art methods for both synthetic and real images.
- 44
- 43
- TMFormer: Token Merging Transformer for Brain Tumor Segmentation with Missing Modalities40
Extensive experimental results on the BraTS 2018 and 2020 datasets demonstrate the superiority and efficacy of TMFormer compared to state-of-the-art methods when dealing with missing modalities.
- EMO-LLaMA: Enhancing Facial Emotion Understanding with Instruction Tuning39
A novel MLLM is proposed, named EMO-LLaMA, which incorporates facial priors from a pretrained facial analysis network to enhance its understanding of human facial information and design a Face Info Mining module to extract both global and local facial information.
- NTIRE 2025 Challenge on Day and Night Raindrop Removal for Dual-Focused Images: Methods and Results37
This paper reviews the NTIRE 2025 Challenge on Day and Night Raindrop Removal for Dual-Focused Images to establish a new and powerful benchmark for the task of removing raindrops under varying lighting and focus conditions.
- 35
- Multi-Scale Promoted Self-Adjusting Correlation Learning for Facial Action Unit Detection32
A novel self-adjusting AU-correlation learning (SACL) method with less computation for AU detection that outperforms the state-of-the-art methods on widely used AU detection benchmark datasets and obtains a more robust feature representation for the final AU detection.
- 32
- 30
- Adversarial Dual-Student With Differentiable Spatial Warping for Semi-Supervised Semantic Segmentation30
A context friendly differentiable geometric warping to conduct unsupervised data augmentation and a novel adversarial dual-student framework is proposed to improve the Mean-Teacher from the following two aspects.
- 30
- 29
- Unsupervised moiré pattern removal for recaptured screen images29
An unsupervised Generative Adversarial Network for Moire Removal (MR-GAN), which is the first attempt at unsuper supervised learning based moire removal, and outperforms state-of-the-art demoireing methods on a large set of test images.
- 29
- 28
- Unsupervised HDR Image and Video Tone Mapping via Contrastive Learning27
This work proposes a unified framework (IVTMNet) for unsupervised image and video tone mapping, and proposes a novel latent code, which is an aggregation of the brightness and contrast of extracted features, to measure the similarity of different pairs.
- 27
- Low‐light image enhancement based on Retinex decomposition and adaptive gamma correction26
Quantitative and qualitative experiments demonstrate the superiority of the proposed low-light image enhancement method over state-of-the-art image enhancement methods.
- 24
- 24
- Recaptured Raw Screen Image and Video Demoiréing via Channel and Spatial Modulations24
This work introduces a color-separated feature branch, and it is fused with the traditional feature-mixed branch via channel and spatial modulations, and achieves state-of-the-art performance for both image and video demori\'eing.
- Raw Image Reconstruction From Rgb on Smartphones. Ntire 2025 Challenge Report23
This paper aims to recover RAW sensor images from smartphones given the corresponding sRGB images without metadata and, by doing this, “reverse” the ISP transformation.
- 23
- 23
- Learning to See Low-Light Images via Feature Domain Adaptation22
A single-stage network empowered by Feature Domain Adaptation (FDA) to decouple the denoising and color mapping tasks in raw LLIE achieves state-of-the-art performance with fewer computing costs (60% FLOPs of the two-stage method DNF).
- 22
- Deep edge map guided depth super resolution22
A deep edge map guided depth SR method, which includes an edge prediction subnetwork and an SR subnetwork, which takes advantage of the hierarchical representation of color and depth images to produce accurate edge maps, which promote the performance of SR sub network.
- RViDeformer: Efficient Raw Video Denoising Transformer With a Larger Benchmark Dataset21
This work proposes an efficient raw video denoising transformer network (RViDeformer) that explores both short and long-distance correlations and employs reparameterization to reduce computation costs.
- Reference-Based Speech Enhancement via Feature Alignment and Fusion Network19
This work proposes reference-based speech enhancement via a feature alignment and fusion network (FAF-Net) and demonstrates that the performance of the proposed FAF-Net is close to state-of-the-art speech enhancement methods on both DNS and Voice Bank+DEMAND datasets.
- 19
- 19
- Automatic Gauge Detection via Geometric Fitting for Safety Inspection18
A gauge detection algorithm based on the methodology of geometric fitting under the framework of random sample consensus (RANSAC) to remove straight lines that do not belong to gauges from captured images.
- FEALLM: Advancing Facial Emotion Analysis in Multimodal Large Language Models with Emotional Synergy and Reasoning15
A novel FEA Instruction Dataset is introduced that provides accurate and aligned FE and AU descriptions and establishes causal reasoning relationships between them, followed by constructing a new benchmark, FEABench, and a novel MLLM architecture is proposed, FEALLM, a novel MLLM architecture designed to capture more detailed facial information, enhancing its capability in FEA tasks.
- 15
- Spatiotemporally scalable matrix recovery for background modeling and moving object detection14
A unified framework called spatiotemporally scalable matrix recovery (SSMR), which has a moderate computational and space complexity scalable to temporal and spatial resolution of videos, and a pyramidal scheme to achieve spatial scalability for high definition videos.
- 14
- 12
- 12
- IMShare11
The proposed IMShare is the first scheme to demonstrate that not only visually pleasant images can be reconstructed using this mobile-cloud method but also the reconstruction can be done in seconds using parallel computing.
- 10
- 9
- 9
- 9
- DSDNet: Raw Domain Demoiréing via Dual Color-Space Synergy8
A single-stage raw domain demoiréing framework, Dual-Stream Demoiréing Network (DSDNet), which leverages the synergy of raw and YCbCr images to remove moiré while preserving luminance and color fidelity and develops a Luminance-Chrominance Adaptive Transformer (LCAT), which decouples luminance and chrominance representations.
- Learning to Reconstruct and Understand Indoor Scenes From Sparse Views8
This paper proposes a new method for simultaneous 3D reconstruction and semantic segmentation for indoor scenes that only needs a small number of color images from uncalibrated sparse views, and proposes a joint global and local registration method to reconstruct a 3D scene with semantic information.
- Efficient HDR Reconstruction From Real-World Raw Images7
This work proposes RepUNet, a lightweight and efficient HDR network leveraging structural re-parameterization for fast and robust inference, and designs a new computational raw HDR data formation pipeline and construct a new raw HDR dataset, RealRaw-HDR.
- Learning Adaptive Lighting via Channel-Aware Guidance7
This work introduces the channel-aware Learning Adaptive Lighting Network (LALNet), a multi-task framework designed to handle multiple light-related tasks efficiently and significantly outperforms state-of-the-art methods on benchmark tests and requires fewer computational resources.
- Reference guided image super-resolution via efficient dense warping and adaptive fusion7
This work proposes an efficient dense warping method to deal with large displacements, which is much faster than traditional patch (or texture) matching strategy and outperforms state-of-the-art SR methods in both subjective and objective measurements.
- 7
- HDR Video Reconstruction with a Large Dynamic Dataset in Raw and sRGB Domains6
Experimental results demonstrate that the proposed dataset can improve the HDR reconstruction performance on real scenes for three benchmark networks and the proposed Raw-HDRNet, which utilizes the raw LDR frames as inputs, is a strong baseline for raw HDR reconstruction.
- 6
- 6
- 6
- 5
- From Recognition to Prediction: Leveraging Sequence Reasoning for Action Anticipation5
A novel end-to-end video modeling architecture that utilizes attention mechanisms, named Anticipation via Recognition and Reasoning (ARR), which decomposes the action anticipation task into action recognition and sequence reasoning tasks and effectively learns the statistical relationship between actions by next action prediction (NAP).
- DeeDSR: Towards Real-World Image Super-Resolution via Degradation-Aware Stable Diffusion5
A novel two-stage, degradation-aware framework that enhances the diffusion model's ability to recognize content and degradation in low-resolution images and recovers semantically precise and photorealistic details, particularly under significant degradation conditions, demonstrating state-of-the-art performance across various benchmarks.
- 5
- ITSRN++: Stronger and Better Implicit Transformer Network for Continuous Screen Content Image Super-Resolution5
An implicit transformer network for continuous SCI SR (termed as ITSRN++) is proposed, which modulates the pixel features in discrete space via a periodic nonlinear function to generate features for continuous pixels.
- Image denoising using cloud images5
A novel image denoising method, which restores an image by exploiting the correlations between the noisy image and the images retrieved from the cloud and achieves significantly better results compared to state-of-the-art methods in terms of both objective and subjective qualities.
- 4
- 4
- 4
- 4
- Distribution-Specific Learning for Joint Salient and Camouflaged Object Detection3
The proposed SCJoint, a joint learning scheme for SOD and COD tasks, and a saliency-based sampling strategy (SBSS) to sample the training set of the SOD task to balance the training set sizes of the two tasks, improve the training set quality and shortens the training time.
- CameraMaster: Unified Camera Semantic-Parameter Control for Photography Retouching3
The key idea is to explicitly decouple the camera directive and then coherently integrate two critical information streams: a directive representation that captures the photographer's intent, and a parameter embedding that encodes precise camera settings.
- 3
- Efficient Screen Content Image Compression via Superpixel-based Content Aggregation and Dynamic Feature Fusion3
A Superpixel-based Content Aggregation Block (SCAB) is proposed to aggregate local pixels into one super-pixel and aggregate non-local information via super-pixel transformer, enabling the dynamic assimilation of non-local information while maintaining manageable complexity.
- Efficient textile anomaly detection via memory guided distillation network3
An efficient memory guided distillation network, which includes encoder, decoder, and segmentation networks, is proposed, which utilizes a small feature extraction network, whose features are distilled from a teacher network to improve the reconstruction quality with small networks.
- Adversarial Robustness in RGB-Skeleton Action Recognition: Leveraging Attention Modality Reweighter3
The Attention-based Modality Reweighter (AMR), which utilizes an attention layer to re-weight the two modalities, enabling the model to learn more robust features, is proposed, which effectively balances the differences in robustness between different modalities.
- 3
- RISAM: Referring Image Segmentation via Mutual-Aware Attention Features3
A mutual-aware attention mechanism consists of Vision-Guided Attention and Language-Guided Attention, which bidirectionally model the relationship between visual and linguistic features and a Mask Decoder to enable explicit linguistic guidance for more consistent segmentation with the language expression.
- 3
- Reference Image Guided Super-Resolution via Progressive Channel Attention Networks3
Extensive quantitative and qualitative evaluations on three benchmark datasets demonstrate that the proposed progressive channel attention network (PCANet) is superior to the state-of-the-art SISR and RefSR methods in terms of PSNR (Peak Signal-to-Noise Ratio) and SSIM (Structural Similarity).
- 3
- F2HDR: Two-Stage HDR Video Reconstruction via Flow Adapter and Physical Motion Modeling2
F2HDR is proposed, a two-stage HDR video reconstruction framework that robustly perceives inter-frame motion and restores fine details in complex dynamic scenarios, and achieves state-of-the-art performance on real-world HDR video benchmarks.
- 2
- 2
- Advances of low-level vision reconstruction in raw domain2
This review focuses on the low - level vision reconstruction technology in the raw domain and mainly investigates the progress of deep learning - based vision reconstruction.
- 2
- Zero-Shot Video Restoration and Enhancement with Assistance of Video Diffusion Models1
This paper proposes the first framework that utilizes the rapidly-developed video diffusion model to assist the image-based method in maintaining more temporal consistency for zero-shot video restoration and enhancement.
- 1
- 1
- rPPG-MAE: Self-supervised Pre-training with Masked Autoencoders for Remote Physiological Measurement1
- 1
- –
- –
- –
- –
- SpiralDiff: Spiral Diffusion with LoRA for RGB-to-RAW Conversion Across Cameras–
SpiralDiff, a diffusion-based framework tailored for RGB-to-RAW conversion with a signal-dependent noise weighting strategy that adapts reconstruction fidelity across intensity levels, is proposed and CamLoRA, a camera-aware lightweight adaptation module that enables a unified model to adapt to different camera-specific ISP characteristics is introduced.
- –
- –
- –
- –
- –
- –
- Deep noise estimation and removal for real-world noisy images–
This paper proposes a joint noise estimation and removal network for real-world noisy images and proposes a RSD (Residual SE D-Conv) block, which explores the spatial and channel correlations of realistic noise.
- –
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 2026-10-12. 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.