Kangfu Mei

Nithin Gopalakrishnan Nair, Kangfu Mei, Vishal M. Patel

Although many long-range imaging systems are designed to support extended vision applications, a natural obstacle to their operation is degradation due to atmospheric turbulence. Atmospheric turbulence causes significant degradation to image quality by introducing blur and geometric distortion. In recent years, various deep learning-based single image atmospheric turbulence mitigation methods, including CNN-based and GAN inversion-based, have been proposed in the literature which attempt to remove the distortion in the image. However, some of these methods are difficult to train and often fail to reconstruct facial features and produce unrealistic results especially in the case of high turbulence. Denoising Diffusion Probabilistic Models (DDPMs) have recently gained some traction because of their stable training process and their ability to generate high quality images. In this paper, we propose the first DDPM-based solution for the problem of atmospheric turbulence mitigation. We also propose a fast sampling technique for reducing the inference times for conditional DDPMs. Extensive experiments are conducted on synthetic and real-world data to show the significance of our model. To facilitate further research, all codes and pretrained models are publically available at http://github.com/Nithin-GK/AT-DDPM

Kangfu Mei, Vishal M. Patel

Diffusion models have emerged as a powerful generative method for synthesizing high-quality and diverse set of images. In this paper, we propose a video generation method based on diffusion models, where the effects of motion are modeled in an implicit condition manner, i.e. one can sample plausible video motions according to the latent feature of frames. We improve the quality of the generated videos by proposing multiple strategies such as sampling space truncation, robustness penalty, and positional group normalization. Various experiments are conducted on datasets consisting of videos with different resolutions and different number of frames. Results show that the proposed method outperforms the state-of-the-art generative adversarial network-based methods by a significant margin in terms of FVD scores as well as perceptible visual quality.

Kangfu Mei, Mauricio Delbracio, Hossein Talebi et al.

Large generative diffusion models have revolutionized text-to-image generation and offer immense potential for conditional generation tasks such as image enhancement, restoration, editing, and compositing. However, their widespread adoption is hindered by the high computational cost, which limits their real-time application. To address this challenge, we introduce a novel method dubbed CoDi, that adapts a pre-trained latent diffusion model to accept additional image conditioning inputs while significantly reducing the sampling steps required to achieve high-quality results. Our method can leverage architectures such as ControlNet to incorporate conditioning inputs without compromising the model's prior knowledge gained during large scale pre-training. Additionally, a conditional consistency loss enforces consistent predictions across diffusion steps, effectively compelling the model to generate high-quality images with conditions in a few steps. Our conditional-task learning and distillation approach outperforms previous distillation methods, achieving a new state-of-the-art in producing high-quality images with very few steps (e.g., 1-4) across multiple tasks, including super-resolution, text-guided image editing, and depth-to-image generation.

Nithin Gopalakrishnan Nair, Kangfu Mei, Vishal M. Patel

Atmospheric turbulence deteriorates the quality of images captured by long-range imaging systems by introducing blur and geometric distortions to the captured scene. This leads to a drastic drop in performance when computer vision algorithms like object/face recognition and detection are performed on these images. In recent years, various deep learning-based atmospheric turbulence mitigation methods have been proposed in the literature. These methods are often trained using synthetically generated images and tested on real-world images. Hence, the performance of these restoration methods depends on the type of simulation used for training the network. In this paper, we systematically evaluate the effectiveness of various turbulence simulation methods on image restoration. In particular, we evaluate the performance of two state-or-the-art restoration networks using six simulations method on a real-world LRFID dataset consisting of face images degraded by turbulence. This paper will provide guidance to the researchers and practitioners working in this field to choose the suitable data generation models for training deep models for turbulence mitigation. The implementation codes for the simulation methods, source codes for the networks, and the pre-trained models will be publicly made available.

Kangfu Mei, Vishal M. Patel, Rui Huang

The ultimate aim of image restoration like denoising is to find an exact correlation between the noisy and clear image domains. But the optimization of end-to-end denoising learning like pixel-wise losses is performed in a sample-to-sample manner, which ignores the intrinsic correlation of images, especially semantics. In this paper, we introduce the Deep Semantic Statistics Matching (D2SM) Denoising Network. It exploits semantic features of pretrained classification networks, then it implicitly matches the probabilistic distribution of clear images at the semantic feature space. By learning to preserve the semantic distribution of denoised images, we empirically find our method significantly improves the denoising capabilities of networks, and the denoised results can be better understood by high-level vision tasks. Comprehensive experiments conducted on the noisy Cityscapes dataset demonstrate the superiority of our method on both the denoising performance and semantic segmentation accuracy. Moreover, the performance improvement observed on our extended tasks including super-resolution and dehazing experiments shows its potentiality as a new general plug-and-play component.

Kangfu Mei, Nithin Gopalakrishnan Nair, Vishal M. Patel

Conditional diffusion probabilistic models can model the distribution of natural images and can generate diverse and realistic samples based on given conditions. However, oftentimes their results can be unrealistic with observable color shifts and textures. We believe that this issue results from the divergence between the probabilistic distribution learned by the model and the distribution of natural images. The delicate conditions gradually enlarge the divergence during each sampling timestep. To address this issue, we introduce a new method that brings the predicted samples to the training data manifold using a pretrained unconditional diffusion model. The unconditional model acts as a regularizer and reduces the divergence introduced by the conditional model at each sampling step. We perform comprehensive experiments to demonstrate the effectiveness of our approach on super-resolution, colorization, turbulence removal, and image-deraining tasks. The improvements obtained by our method suggest that the priors can be incorporated as a general plugin for improving conditional diffusion models.

Kangfu Mei, Yiqun Mei, Vishal M. Patel

In many practical applications of long-range imaging such as biometrics and surveillance, thermal imagining modalities are often used to capture images in low-light and nighttime conditions. However, such imaging systems often suffer from atmospheric turbulence, which introduces severe blur and deformation artifacts to the captured images. Such an issue is unavoidable in long-range imaging and significantly decreases the face verification accuracy. In this paper, we first investigate the problem with a turbulence simulation method on real-world thermal images. An end-to-end reconstruction method is then proposed which can directly transform thermal images into visible-spectrum images by utilizing natural image priors based on a pre-trained StyleGAN2 network. Compared with the existing two-steps methods of consecutive turbulence mitigation and thermal to visible image translation, our method is demonstrated to be effective in terms of both the visual quality of the reconstructed results and face verification accuracy. Moreover, to the best of our knowledge, this is the first work that studies the problem of thermal to visible image translation under atmospheric turbulence.

Jinxiu Liu, Xuanming Liu, Kangfu Mei et al.

Large pretrained diffusion models have significantly enhanced the quality of generated videos, and yet their use in real-time streaming remains limited. Autoregressive models offer a natural framework for sequential frame synthesis but require heavy computation to achieve high fidelity. Diffusion distillation can compress these models into efficient few-step variants, but existing video distillation approaches largely adapt image-specific methods that neglect temporal dependencies. These techniques often excel in image generation but underperform in video synthesis, exhibiting reduced motion coherence, error accumulation over long sequences, and a latency-quality trade-off. We identify two factors that result in these limitations: insufficient utilization of temporal context during step reduction and implicit prediction of subsequent noise levels in next-chunk prediction (i.e., exposure bias). To address these issues, we propose Diagonal Distillation, which operates orthogonally to existing approaches and better exploits temporal information across both video chunks and denoising steps. Central to our approach is an asymmetric generation strategy: more steps early, fewer steps later. This design allows later chunks to inherit rich appearance information from thoroughly processed early chunks, while using partially denoised chunks as conditional inputs for subsequent synthesis. By aligning the implicit prediction of subsequent noise levels during chunk generation with the actual inference conditions, our approach mitigates error propagation and reduces oversaturation in long-range sequences. We further incorporate implicit optical flow modeling to preserve motion quality under strict step constraints. Our method generates a 5-second video in 2.61 seconds (up to 31 FPS), achieving a 277.3x speedup over the undistilled model.

Juncheng Li, Faming Fang, Jiaqian Li et al.

Convolutional neural networks have been proven to be of great benefit for single-image super-resolution (SISR). However, previous works do not make full use of multi-scale features and ignore the inter-scale correlation between different upsampling factors, resulting in sub-optimal performance. Instead of blindly increasing the depth of the network, we are committed to mining image features and learning the inter-scale correlation between different upsampling factors. To achieve this, we propose a Multi-scale Dense Cross Network (MDCN), which achieves great performance with fewer parameters and less execution time. MDCN consists of multi-scale dense cross blocks (MDCBs), hierarchical feature distillation block (HFDB), and dynamic reconstruction block (DRB). Among them, MDCB aims to detect multi-scale features and maximize the use of image features flow at different scales, HFDB focuses on adaptively recalibrate channel-wise feature responses to achieve feature distillation, and DRB attempts to reconstruct SR images with different upsampling factors in a single model. It is worth noting that all these modules can run independently. It means that these modules can be selectively plugged into any CNN model to improve model performance. Extensive experiments show that MDCN achieves competitive results in SISR, especially in the reconstruction task with multiple upsampling factors. The code will be provided at https://github.com/MIVRC/MDCN-PyTorch.

Kangfu Mei, Juncheng Li, Jiajie Zhang et al.

Neural-networks based image restoration methods tend to use low-resolution image patches for training. Although higher-resolution image patches can provide more global information, state-of-the-art methods cannot utilize them due to their huge GPU memory usage, as well as the instable training process. However, plenty of studies have shown that global information is crucial for image restoration tasks like image demosaicing and enhancing. In this work, we propose a HighEr-Resolution Network (HERN) to fully learning global information in high-resolution image patches. To achieve this, the HERN employs two parallel paths to learn image features in two different resolutions, respectively. By combining global-aware features and multi-scale features, our HERN is able to learn global information with feasible GPU memory usage. Besides, we introduce a progressive training method to solve the instability issue and accelerate model convergence. On the task of image demosaicing and enhancing, our HERN achieves state-of-the-art performance on the AIM2019 RAW to RGB mapping challenge. The source code of our implementation is available at https://github.com/MKFMIKU/RAW2RGBNet.

Kangfu Mei, Luis Figueroa, Zhe Lin et al.

Recovering textures under shadows has remained a challenging problem due to the difficulty of inferring shadow-free scenes from shadow images. In this paper, we propose the use of diffusion models as they offer a promising approach to gradually refine the details of shadow regions during the diffusion process. Our method improves this process by conditioning on a learned latent feature space that inherits the characteristics of shadow-free images, thus avoiding the limitation of conventional methods that condition on degraded images only. Additionally, we propose to alleviate potential local optima during training by fusing noise features with the diffusion network. We demonstrate the effectiveness of our approach which outperforms the previous best method by 13% in terms of RMSE on the AISTD dataset. Further, we explore instance-level shadow removal, where our model outperforms the previous best method by 82% in terms of RMSE on the DESOBA dataset.

Kangfu Mei, Zhengzhong Tu, Mauricio Delbracio et al.

We study the scaling properties of latent diffusion models (LDMs) with an emphasis on their sampling efficiency. While improved network architecture and inference algorithms have shown to effectively boost sampling efficiency of diffusion models, the role of model size -- a critical determinant of sampling efficiency -- has not been thoroughly examined. Through empirical analysis of established text-to-image diffusion models, we conduct an in-depth investigation into how model size influences sampling efficiency across varying sampling steps. Our findings unveil a surprising trend: when operating under a given inference budget, smaller models frequently outperform their larger equivalents in generating high-quality results. Moreover, we extend our study to demonstrate the generalizability of the these findings by applying various diffusion samplers, exploring diverse downstream tasks, evaluating post-distilled models, as well as comparing performance relative to training compute. These findings open up new pathways for the development of LDM scaling strategies which can be employed to enhance generative capabilities within limited inference budgets.

Kangfu Mei, Hossein Talebi, Mojtaba Ardakani et al.

Single-image super-resolution (SISR) remains challenging due to the inherent difficulty of recovering fine-grained details and preserving perceptual quality from low-resolution inputs. Existing methods often rely on limited image priors, leading to suboptimal results. We propose a novel approach that leverages the rich contextual information available in multiple modalities -- including depth, segmentation, edges, and text prompts -- to learn a powerful generative prior for SISR within a diffusion model framework. We introduce a flexible network architecture that effectively fuses multimodal information, accommodating an arbitrary number of input modalities without requiring significant modifications to the diffusion process. Crucially, we mitigate hallucinations, often introduced by text prompts, by using spatial information from other modalities to guide regional text-based conditioning. Each modality's guidance strength can also be controlled independently, allowing steering outputs toward different directions, such as increasing bokeh through depth or adjusting object prominence via segmentation. Extensive experiments demonstrate that our model surpasses state-of-the-art generative SISR methods, achieving superior visual quality and fidelity. See project page at https://mmsr.kfmei.com/.

Yuyang Hu, Kangfu Mei, Mojtaba Sahraee-Ardakan et al.

Diffusion models show promise for image restoration, but existing methods often struggle with inconsistent fidelity and undesirable artifacts. To address this, we introduce Kernel Density Steering (KDS), a novel inference-time framework promoting robust, high-fidelity outputs through explicit local mode-seeking. KDS employs an $N$-particle ensemble of diffusion samples, computing patch-wise kernel density estimation gradients from their collective outputs. These gradients steer patches in each particle towards shared, higher-density regions identified within the ensemble. This collective local mode-seeking mechanism, acting as "collective wisdom", steers samples away from spurious modes prone to artifacts, arising from independent sampling or model imperfections, and towards more robust, high-fidelity structures. This allows us to obtain better quality samples at the expense of higher compute by simultaneously sampling multiple particles. As a plug-and-play framework, KDS requires no retraining or external verifiers, seamlessly integrating with various diffusion samplers. Extensive numerical validations demonstrate KDS substantially improves both quantitative and qualitative performance on challenging real-world super-resolution and image inpainting tasks.

Yuyang Hu, Mojtaba Sahraee-Ardakan, Arpit Bansal et al.

Pre-trained image restoration models often fail on real-world, out-of-distribution degradations due to significant domain gaps. Adapting to these unseen domains is challenging, as out-of-distribution data lacks ground truth, and traditional adaptation methods often require complex architectural changes. We propose LEGO (Learning from a Generative Oracle), a practical three-stage framework for post-training domain adaptation without paired data. LEGO converts this unsupervised challenge into a tractable pseudo-supervised one. First, we obtain initial restorations from the pre-trained model. Second, we leverage a frozen, large-scale generative oracle to refine these estimates into high-quality pseudo-ground-truths. Third, we fine-tune the original model using a mixed-supervision strategy combining in-distribution data with these new pseudo-pairs. This approach adapts the model to the new distribution without sacrificing its original robustness or requiring architectural modifications. Experiments demonstrate that LEGO effectively bridges the domain gap, significantly improving performance on diverse real-world benchmarks.

Mo Zhou, Keren Ye, Viraj Shah et al.

Preserving face identity is a critical yet persistent challenge in diffusion-based image restoration. While reference faces offer a path forward, existing reference-based methods often fail to fully exploit their potential. This paper introduces a novel approach that maximizes reference face utility for improved face restoration and identity preservation. Our method makes three key contributions: 1) Composite Context, a comprehensive representation that fuses multi-level (high- and low-level) information from the reference face, offering richer guidance than prior singular representations. 2) Hard Example Identity Loss, a novel loss function that leverages the reference face to address the identity learning inefficiencies found in the existing identity loss. 3) A training-free method to adapt the model to multi-reference inputs during inference. The proposed method demonstrably restores high-quality faces and achieves state-of-the-art identity preserving restoration on benchmarks such as FFHQ-Ref and CelebA-Ref-Test, consistently outperforming previous work.

Kangfu Mei, Mo Zhou, Vishal M. Patel

Diffusion Probabilistic Field (DPF) models the distribution of continuous functions defined over metric spaces. While DPF shows great potential for unifying data generation of various modalities including images, videos, and 3D geometry, it does not scale to a higher data resolution. This can be attributed to the ``scaling property'', where it is difficult for the model to capture local structures through uniform sampling. To this end, we propose a new model comprising of a view-wise sampling algorithm to focus on local structure learning, and incorporating additional guidance, e.g., text description, to complement the global geometry. The model can be scaled to generate high-resolution data while unifying multiple modalities. Experimental results on data generation in various modalities demonstrate the effectiveness of our model, as well as its potential as a foundation framework for scalable modality-unified visual content generation.

Kangfu Mei, Vishal M. Patel

In many applications of long-range imaging, we are faced with a scenario where a person appearing in the captured imagery is often degraded by atmospheric turbulence. However, restoring such degraded images for face verification is difficult since the degradation causes images to be geometrically distorted and blurry. To mitigate the turbulence effect, in this paper, we propose the first turbulence mitigation method that makes use of visual priors encapsulated by a well-trained GAN. Based on the visual priors, we propose to learn to preserve the identity of restored images on a spatial periodic contextual distance. Such a distance can keep the realism of restored images from the GAN while considering the identity difference at the network learning. In addition, hierarchical pseudo connections are proposed for facilitating the identity-preserving learning by introducing more appearance variance without identity changing. Extensive experiments show that our method significantly outperforms prior art in both the visual quality and face verification accuracy of restored results.

Kangfu Mei, Shenglong Ye, Rui Huang

Deep Neural Network (DNN) based super-resolution algorithms have greatly improved the quality of the generated images. However, these algorithms often yield significant artifacts when dealing with real-world super-resolution problems due to the difficulty in learning misaligned optical zoom. In this paper, we introduce a Squared Deformable Alignment Network (SDAN) to address this issue. Our network learns squared per-point offsets for convolutional kernels, and then aligns features in corrected convolutional windows based on the offsets. So the misalignment will be minimized by the extracted aligned features. Different from the per-point offsets used in the vanilla Deformable Convolutional Network (DCN), our proposed squared offsets not only accelerate the offset learning but also improve the generation quality with fewer parameters. Besides, we further propose an efficient cross packing attention layer to boost the accuracy of the learned offsets. It leverages the packing and unpacking operations to enlarge the receptive field of the offset learning and to enhance the ability of extracting the spatial connection between the low-resolution images and the referenced images. Comprehensive experiments show the superiority of our method over other state-of-the-art methods in both computational efficiency and realistic details.

Qi Song, Kangfu Mei, Rui Huang

Two factors have proven to be very important to the performance of semantic segmentation models: global context and multi-level semantics. However, generating features that capture both factors always leads to high computational complexity, which is problematic in real-time scenarios. In this paper, we propose a new model, called Attention-Augmented Network (AttaNet), to capture both global context and multilevel semantics while keeping the efficiency high. AttaNet consists of two primary modules: Strip Attention Module (SAM) and Attention Fusion Module (AFM). Viewing that in challenging images with low segmentation accuracy, there are a significantly larger amount of vertical strip areas than horizontal ones, SAM utilizes a striping operation to reduce the complexity of encoding global context in the vertical direction drastically while keeping most of contextual information, compared to the non-local approaches. Moreover, AFM follows a cross-level aggregation strategy to limit the computation, and adopts an attention strategy to weight the importance of different levels of features at each pixel when fusing them, obtaining an efficient multi-level representation. We have conducted extensive experiments on two semantic segmentation benchmarks, and our network achieves different levels of speed/accuracy trade-offs on Cityscapes, e.g., 71 FPS/79.9% mIoU, 130 FPS/78.5% mIoU, and 180 FPS/70.1% mIoU, and leading performance on ADE20K as well.

Qiaosi Yi, Yunxing Liu, Aiwen Jiang et al.

Crowd counting is an important task that shown great application value in public safety-related fields, which has attracted increasing attention in recent years. In the current research, the accuracy of counting numbers and crowd density estimation are the main concerns. Although the emergence of deep learning has greatly promoted the development of this field, crowd counting under cluttered background is still a serious challenge. In order to solve this problem, we propose a ScaleAware Crowd Counting Network (SACCN) with regional and semantic attentions. The proposed SACCN distinguishes crowd and background by applying regional and semantic self-attention mechanisms on the shallow layers and deep layers, respectively. Moreover, the asymmetric multi-scale module (AMM) is proposed to deal with the problem of scale diversity, and regional attention based dense connections and skip connections are designed to alleviate the variations on crowd scales. Extensive experimental results on multiple public benchmarks demonstrate that our proposed SACCN achieves satisfied superior performances and outperform most state-of-the-art methods. All codes and pretrained models will be released soon.

Kangfu Mei, Yao Lu, Qiaosi Yi et al.

Pursuing realistic results according to human visual perception is the central concern in the image transformation tasks. Perceptual learning approaches like perceptual loss are empirically powerful for such tasks but they usually rely on the pre-trained classification network to provide features, which are not necessarily optimal in terms of visual perception of image transformation. In this paper, we argue that, among the features representation from the pre-trained classification network, only limited dimensions are related to human visual perception, while others are irrelevant, although both will affect the final image transformation results. Under such an assumption, we try to disentangle the perception-relevant dimensions from the representation through our proposed online contrastive learning. The resulted network includes the pre-training part and a feature selection layer, followed by the contrastive learning module, which utilizes the transformed results, target images, and task-oriented distorted images as the positive, negative, and anchor samples, respectively. The contrastive learning aims at activating the perception-relevant dimensions and suppressing the irrelevant ones by using the triplet loss, so that the original representation can be disentangled for better perceptual quality. Experiments on various image transformation tasks demonstrate the superiority of our framework, in terms of human visual perception, to the existing approaches using pre-trained networks and empirically designed losses.

Kangfu Mei, Aiwen Jiang, Juncheng Li et al.

Single image dehazing is a challenging ill-posed restoration problem. Various prior-based and learning-based methods have been proposed. Most of them follow a classic atmospheric scattering model which is an elegant simplified physical model based on the assumption of single-scattering and homogeneous atmospheric medium. The formulation of haze in realistic environment is more complicated. In this paper, we propose to take its essential mechanism as "black box", and focus on learning an input-adaptive trainable end-to-end dehazing model. An U-Net like encoder-decoder deep network via progressive feature fusions has been proposed to directly learn highly nonlinear transformation function from observed hazy image to haze-free ground-truth. The proposed network is evaluated on two public image dehazing benchmarks. The experiments demonstrate that it can achieve superior performance when compared with popular state-of-the-art methods. With efficient GPU memory usage, it can satisfactorily recover ultra high definition hazed image up to 4K resolution, which is unaffordable by many deep learning based dehazing algorithms.

Kangfu Mei, Aiwen Jiang, Juncheng Li et al.

Recent works on single-image super-resolution are concentrated on improving performance through enhancing spatial encoding between convolutional layers. In this paper, we focus on modeling the correlations between channels of convolutional features. We present an effective deep residual network based on squeeze-and-excitation blocks (SEBlock) to reconstruct high-resolution (HR) image from low-resolution (LR) image. SEBlock is used to adaptively recalibrate channel-wise feature mappings. Further, short connections between each SEBlock are used to remedy information loss. Extensive experiments show that our model can achieve the state-of-the-art performance and get finer texture details.