Pub Date : 2024-10-22DOI: 10.1109/TIP.2024.3482175
Yinsong Xu;Jiaqi Tang;Aidong Men;Qingchao Chen
Medical image segmentation is a critical task in clinical applications. Recently, the Segment Anything Model (SAM) has demonstrated potential for natural image segmentation. However, the requirement for expert labour to provide prompts, and the domain gap between natural and medical images pose significant obstacles in adapting SAM to medical images. To overcome these challenges, this paper introduces a novel prompt generation method named EviPrompt. The proposed method requires only a single reference image-annotation pair, making it a training-free solution that significantly reduces the need for extensive labelling and computational resources. First, prompts are automatically generated based on the similarity between features of the reference and target images, and evidential learning is introduced to improve reliability. Then, to mitigate the impact of the domain gap, committee voting and inference-guided in-context learning are employed, generating prompts primarily based on human prior knowledge and reducing reliance on extracted semantic information. EviPrompt represents an efficient and robust approach to medical image segmentation. We evaluate it across a broad range of tasks and modalities, confirming its efficacy. The source code is available at https://github.com/SPIresearch/EviPrompt .
医学图像分割是临床应用中的一项关键任务。最近，"任意分割模型"（Segment Anything Model，SAM）在自然图像分割方面展现出了潜力。然而，由于需要专家提供提示，而且自然图像与医学图像之间存在领域差距，因此将 SAM 应用于医学图像存在重大障碍。为了克服这些挑战，本文介绍了一种名为 EviPrompt 的新型提示生成方法。所提出的方法只需要一个参考图像-注释对，是一种无需训练的解决方案，大大减少了对大量标注和计算资源的需求。首先，根据参考图像和目标图像特征之间的相似性自动生成提示，并引入证据学习以提高可靠性。然后，为了减轻领域差距的影响，采用了委员会投票和推理引导的上下文学习，主要根据人类的先验知识生成提示，减少对提取的语义信息的依赖。EviPrompt 是一种高效、稳健的医学图像分割方法。我们在广泛的任务和模式中对其进行了评估，证实了它的功效。源代码见 https://github.com/SPIresearch/EviPrompt。
{"title":"EviPrompt: A Training-Free Evidential Prompt Generation Method for Adapting Segment Anything Model in Medical Images","authors":"Yinsong Xu;Jiaqi Tang;Aidong Men;Qingchao Chen","doi":"10.1109/TIP.2024.3482175","DOIUrl":"10.1109/TIP.2024.3482175","url":null,"abstract":"Medical image segmentation is a critical task in clinical applications. Recently, the Segment Anything Model (SAM) has demonstrated potential for natural image segmentation. However, the requirement for expert labour to provide prompts, and the domain gap between natural and medical images pose significant obstacles in adapting SAM to medical images. To overcome these challenges, this paper introduces a novel prompt generation method named EviPrompt. The proposed method requires only a single reference image-annotation pair, making it a training-free solution that significantly reduces the need for extensive labelling and computational resources. First, prompts are automatically generated based on the similarity between features of the reference and target images, and evidential learning is introduced to improve reliability. Then, to mitigate the impact of the domain gap, committee voting and inference-guided in-context learning are employed, generating prompts primarily based on human prior knowledge and reducing reliance on extracted semantic information. EviPrompt represents an efficient and robust approach to medical image segmentation. We evaluate it across a broad range of tasks and modalities, confirming its efficacy. The source code is available at \u0000<uri>https://github.com/SPIresearch/EviPrompt</uri>\u0000.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6204-6215"},"PeriodicalIF":0.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142487459","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1109/TIP.2024.3482185
Kevin Zhang;Sakshum Kulshrestha;Christopher A. Metzler
Despite recent advances, developing general-purpose universal denoising and artifact-removal networks remains largely an open problem: Given fixed network weights, one inherently trades-off specialization at one task (e.g., removing Poisson noise) for performance at another (e.g., removing speckle noise). In addition, training such a network is challenging due to the curse of dimensionality: As one increases the dimensions of the specification-space (i.e., the number of parameters needed to describe the noise distribution) the number of unique specifications one needs to train for grows exponentially. Uniformly sampling this space will result in a network that does well at very challenging problem specifications but poorly at easy problem specifications, where even large errors will have a small effect on the overall mean squared error. In this work we propose training denoising networks using an adaptive-sampling/active-learning strategy. Our work improves upon a recently proposed universal denoiser training strategy by extending these results to higher dimensions and by incorporating a polynomial approximation of the true specification-loss landscape. This approximation allows us to reduce training times by almost two orders of magnitude. We test our method on simulated joint Poisson-Gaussian-Speckle noise and demonstrate that with our proposed training strategy, a single blind, generalist denoiser network can achieve peak signal-to-noise ratios within a uniform bound of specialized denoiser networks across a large range of operating conditions. We also capture a small dataset of images with varying amounts of joint Poisson-Gaussian-Speckle noise and demonstrate that a universal denoiser trained using our adaptive-sampling strategy outperforms uniformly trained baselines.
尽管最近取得了一些进展，但开发通用的通用去噪和去除伪影网络在很大程度上仍然是一个有待解决的问题：在网络权重固定的情况下，人们需要在一项任务（如去除泊松噪声）的专业性与另一项任务（如去除斑点噪声）的性能之间进行权衡。此外，由于 "维度诅咒"（curse of dimensionality）的存在，训练这样的网络具有挑战性：随着规格空间维度（即描述噪声分布所需的参数数量）的增加，需要训练的独特规格数量也呈指数增长。对这一空间进行均匀采样会导致网络在处理极具挑战性的问题规格时表现出色，但在处理简单的问题规格时却表现不佳，在这种情况下，即使误差很大，对总体均方误差的影响也很小。在这项工作中，我们建议使用自适应采样/主动学习策略来训练去噪网络。我们的工作改进了最近提出的通用去噪器训练策略，将这些结果扩展到了更高的维度，并加入了对真实规格损失情况的多项式近似。这种近似方法使我们的训练时间缩短了近两个数量级。我们在模拟泊松-高斯-啄木鸟联合噪声的基础上测试了我们的方法，结果表明，采用我们提出的训练策略，单个盲通用去噪网络可以在大范围的操作条件下，在专用去噪网络的统一范围内达到峰值信噪比。我们还捕捉了一个具有不同数量泊松-高斯-啄木鸟联合噪声的小型图像数据集，并证明使用我们的自适应采样策略训练的通用去噪器优于统一训练的基线。
{"title":"A Scalable Training Strategy for Blind Multi-Distribution Noise Removal","authors":"Kevin Zhang;Sakshum Kulshrestha;Christopher A. Metzler","doi":"10.1109/TIP.2024.3482185","DOIUrl":"10.1109/TIP.2024.3482185","url":null,"abstract":"Despite recent advances, developing general-purpose universal denoising and artifact-removal networks remains largely an open problem: Given fixed network weights, one inherently trades-off specialization at one task (e.g., removing Poisson noise) for performance at another (e.g., removing speckle noise). In addition, training such a network is challenging due to the curse of dimensionality: As one increases the dimensions of the specification-space (i.e., the number of parameters needed to describe the noise distribution) the number of unique specifications one needs to train for grows exponentially. Uniformly sampling this space will result in a network that does well at very challenging problem specifications but poorly at easy problem specifications, where even large errors will have a small effect on the overall mean squared error. In this work we propose training denoising networks using an adaptive-sampling/active-learning strategy. Our work improves upon a recently proposed universal denoiser training strategy by extending these results to higher dimensions and by incorporating a polynomial approximation of the true specification-loss landscape. This approximation allows us to reduce training times by almost two orders of magnitude. We test our method on simulated joint Poisson-Gaussian-Speckle noise and demonstrate that with our proposed training strategy, a single blind, generalist denoiser network can achieve peak signal-to-noise ratios within a uniform bound of specialized denoiser networks across a large range of operating conditions. We also capture a small dataset of images with varying amounts of joint Poisson-Gaussian-Speckle noise and demonstrate that a universal denoiser trained using our adaptive-sampling strategy outperforms uniformly trained baselines.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6216-6226"},"PeriodicalIF":0.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142487456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1109/TIP.2024.3482189
Zexuan Ji;Shunlong Ye;Xiao Ma
Accurate and automatic segmentation of medical images plays an essential role in clinical diagnosis and analysis. It has been established that integrating contextual relationships substantially enhances the representational ability of neural networks. Conventionally, Long Short-Term Memory (LSTM) and Self-Attention (SA) mechanisms have been recognized for their proficiency in capturing global dependencies within data. However, these mechanisms have typically been viewed as distinct modules without a direct linkage. This paper presents the integration of LSTM design with SA sparse coding as a key innovation. It uses linear combinations of LSTM states for SA’s query, key, and value (QKV) matrices to leverage LSTM’s capability for state compression and historical data retention. This approach aims to rectify the shortcomings of conventional sparse coding methods that overlook temporal information, thereby enhancing SA’s ability to do sparse coding and capture global dependencies. Building upon this premise, we introduce two innovative modules that weave the SA matrix into the LSTM state design in distinct manners, enabling LSTM to more adeptly model global dependencies and meld seamlessly with SA without accruing extra computational demands. Both modules are separately embedded into the U-shaped convolutional neural network architecture for handling both 2D and 3D medical images. Experimental evaluations on downstream medical image segmentation tasks reveal that our proposed modules not only excel on four extensively utilized datasets across various baselines but also enhance prediction accuracy, even on baselines that have already incorporated contextual modules. Code is available at https://github.com/yeshunlong/SALSTM .
准确、自动地分割医学图像在临床诊断和分析中发挥着至关重要的作用。已经证实，整合上下文关系可大大增强神经网络的表征能力。传统上，长短时记忆（LSTM）和自我注意（SA）机制因其捕捉数据内全局依赖关系的能力而得到认可。然而，这些机制通常被视为不同的模块，没有直接联系。本文将 LSTM 设计与 SA 稀疏编码相结合，作为一项重要创新。它将 LSTM 状态的线性组合用于 SA 的查询、键和值（QKV）矩阵，以充分利用 LSTM 的状态压缩和历史数据保留能力。这种方法旨在纠正传统稀疏编码方法忽略时间信息的缺点，从而增强 SA 的稀疏编码和捕捉全局依赖性的能力。在这一前提下，我们引入了两个创新模块，它们以不同的方式将 SA 矩阵编织到 LSTM 状态设计中，使 LSTM 能够更巧妙地模拟全局依赖关系，并与 SA 无缝结合，而不会产生额外的计算需求。这两个模块分别嵌入到 U 型卷积神经网络架构中，用于处理二维和三维医学图像。对下游医学图像分割任务的实验评估表明，我们提出的模块不仅在四种广泛使用的数据集上表现出色，而且还提高了预测准确性，即使在已经包含上下文模块的基线上也是如此。代码见 https://github.com/yeshunlong/SALSTM。
{"title":"Sparse Coding Inspired LSTM and Self-Attention Integration for Medical Image Segmentation","authors":"Zexuan Ji;Shunlong Ye;Xiao Ma","doi":"10.1109/TIP.2024.3482189","DOIUrl":"10.1109/TIP.2024.3482189","url":null,"abstract":"Accurate and automatic segmentation of medical images plays an essential role in clinical diagnosis and analysis. It has been established that integrating contextual relationships substantially enhances the representational ability of neural networks. Conventionally, Long Short-Term Memory (LSTM) and Self-Attention (SA) mechanisms have been recognized for their proficiency in capturing global dependencies within data. However, these mechanisms have typically been viewed as distinct modules without a direct linkage. This paper presents the integration of LSTM design with SA sparse coding as a key innovation. It uses linear combinations of LSTM states for SA’s query, key, and value (QKV) matrices to leverage LSTM’s capability for state compression and historical data retention. This approach aims to rectify the shortcomings of conventional sparse coding methods that overlook temporal information, thereby enhancing SA’s ability to do sparse coding and capture global dependencies. Building upon this premise, we introduce two innovative modules that weave the SA matrix into the LSTM state design in distinct manners, enabling LSTM to more adeptly model global dependencies and meld seamlessly with SA without accruing extra computational demands. Both modules are separately embedded into the U-shaped convolutional neural network architecture for handling both 2D and 3D medical images. Experimental evaluations on downstream medical image segmentation tasks reveal that our proposed modules not only excel on four extensively utilized datasets across various baselines but also enhance prediction accuracy, even on baselines that have already incorporated contextual modules. Code is available at \u0000<uri>https://github.com/yeshunlong/SALSTM</uri>\u0000.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6098-6113"},"PeriodicalIF":0.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142487460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The existence of noisy labels in real-world data negatively impacts the performance of deep learning models. Although much research effort has been devoted to improving the robustness towards noisy labels in classification tasks, the problem of noisy labels in deep metric learning (DML) remains under-explored. Existing noisy label learning methods designed for DML mainly discard suspicious noisy samples, resulting in a waste of the training data. To address this issue, we propose a noise-robust DML framework with SubGroup-based Positive-pair Selection (SGPS), which constructs reliable positive pairs for noisy samples to enhance the sample utilization. Specifically, SGPS first effectively identifies clean and noisy samples by a probability-based clean sample selectionstrategy. To further utilize the remaining noisy samples, we discover their potential similar samples based on the subgroup information given by a subgroup generation module and then aggregate them into informative positive prototypes for each noisy sample via a positive prototype generation module. Afterward, a new contrastive loss is tailored for the noisy samples with their selected positive pairs. SGPS can be easily integrated into the training process of existing pair-wise DML tasks, like image retrieval and face recognition. Extensive experiments on multiple synthetic and real-world large-scale label noise datasets demonstrate the effectiveness of our proposed method. Without any bells and whistles, our SGPS framework outperforms the state-of-the-art noisy label DML methods.
{"title":"Enhancing Sample Utilization in Noise-Robust Deep Metric Learning With Subgroup-Based Positive-Pair Selection","authors":"Zhipeng Yu;Qianqian Xu;Yangbangyan Jiang;Yingfei Sun;Qingming Huang","doi":"10.1109/TIP.2024.3482182","DOIUrl":"10.1109/TIP.2024.3482182","url":null,"abstract":"The existence of noisy labels in real-world data negatively impacts the performance of deep learning models. Although much research effort has been devoted to improving the robustness towards noisy labels in classification tasks, the problem of noisy labels in deep metric learning (DML) remains under-explored. Existing noisy label learning methods designed for DML mainly discard suspicious noisy samples, resulting in a waste of the training data. To address this issue, we propose a noise-robust DML framework with SubGroup-based Positive-pair Selection (SGPS), which constructs reliable positive pairs for noisy samples to enhance the sample utilization. Specifically, SGPS first effectively identifies clean and noisy samples by a probability-based clean sample selectionstrategy. To further utilize the remaining noisy samples, we discover their potential similar samples based on the subgroup information given by a subgroup generation module and then aggregate them into informative positive prototypes for each noisy sample via a positive prototype generation module. Afterward, a new contrastive loss is tailored for the noisy samples with their selected positive pairs. SGPS can be easily integrated into the training process of existing pair-wise DML tasks, like image retrieval and face recognition. Extensive experiments on multiple synthetic and real-world large-scale label noise datasets demonstrate the effectiveness of our proposed method. Without any bells and whistles, our SGPS framework outperforms the state-of-the-art noisy label DML methods.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6083-6097"},"PeriodicalIF":0.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142487457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The sphere-to-plane projection of 360-degree video introduces substantial stretched redundant data, which is discarded when reprojected to the 3D sphere for display. Consequently, encoding and transmitting such redundant data is unnecessary. Highly redundant blocks can be referred to as all-zero blocks (AZBs). Detecting these AZBs in advance can reduce computational and transmission resource consumption. However, this cannot be achieved by existing AZB detection techniques due to the unawareness of the stretching redundancy. In this paper, we first derive a latitude-adaptive redundancy detection (LARD) approach to adaptively detect coefficients carrying redundancy in transformed blocks by modeling the dependency between valid frequency range and the stretching degree based on spectrum analysis. Utilizing LARD, a latitude-redundancy-aware AZB detection scheme tailored for fast 360-degree video coding (LRAS) is proposed to accelerate the encoding process. LRAS consists of three sequential stages: latitude-adaptive AZB (L-AZB) detection, latitude-adaptive genuine-AZB (LG-AZB) detection and latitude-adaptive pseudo-AZB (LP-AZB) detection. Specifically, L-AZB refers to the AZB introduced by projection. LARD is used to detect L-AZB directly. LG-AZB refers to the AZB after hard-decision quantization and zeroing redundant coefficients. A novel latitude-adaptive sum of absolute difference estimation model is built to derive the threshold for LG-AZB detection. LP-AZB refers to the AZB in terms of rate-distortion optimization considering redundancy. A latitude-adaptive rate-distortion model is established for LP-AZB detection. Experimental results show that LRAS can achieve an average total encoding time reduction of 25.85% and 20.38% under low-delay and random access configurations compared to the original HEVC encoder, with only 0.16% and 0.13% BDBR increases and 0.01dB BDPSNR loss, respectively. The transform and quantization time savings are 60.13% and 59.94% on average.
{"title":"Latitude-Redundancy-Aware All-Zero Block Detection for Fast 360-Degree Video Coding","authors":"Chang Yu;Xiaopeng Fan;Pengjin Chen;Yuxin Ni;Hengyu Man;Debin Zhao","doi":"10.1109/TIP.2024.3482172","DOIUrl":"10.1109/TIP.2024.3482172","url":null,"abstract":"The sphere-to-plane projection of 360-degree video introduces substantial stretched redundant data, which is discarded when reprojected to the 3D sphere for display. Consequently, encoding and transmitting such redundant data is unnecessary. Highly redundant blocks can be referred to as all-zero blocks (AZBs). Detecting these AZBs in advance can reduce computational and transmission resource consumption. However, this cannot be achieved by existing AZB detection techniques due to the unawareness of the stretching redundancy. In this paper, we first derive a latitude-adaptive redundancy detection (LARD) approach to adaptively detect coefficients carrying redundancy in transformed blocks by modeling the dependency between valid frequency range and the stretching degree based on spectrum analysis. Utilizing LARD, a latitude-redundancy-aware AZB detection scheme tailored for fast 360-degree video coding (LRAS) is proposed to accelerate the encoding process. LRAS consists of three sequential stages: latitude-adaptive AZB (L-AZB) detection, latitude-adaptive genuine-AZB (LG-AZB) detection and latitude-adaptive pseudo-AZB (LP-AZB) detection. Specifically, L-AZB refers to the AZB introduced by projection. LARD is used to detect L-AZB directly. LG-AZB refers to the AZB after hard-decision quantization and zeroing redundant coefficients. A novel latitude-adaptive sum of absolute difference estimation model is built to derive the threshold for LG-AZB detection. LP-AZB refers to the AZB in terms of rate-distortion optimization considering redundancy. A latitude-adaptive rate-distortion model is established for LP-AZB detection. Experimental results show that LRAS can achieve an average total encoding time reduction of 25.85% and 20.38% under low-delay and random access configurations compared to the original HEVC encoder, with only 0.16% and 0.13% BDBR increases and 0.01dB BDPSNR loss, respectively. The transform and quantization time savings are 60.13% and 59.94% on average.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6129-6142"},"PeriodicalIF":0.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142487458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-21DOI: 10.1109/TIP.2024.3480696
Haoyou Deng;Lida Li;Feng Zhang;Zhiqiang Li;Bin Xu;Qingbo Lu;Changxin Gao;Nong Sang
Lens flare is a common phenomenon when strong light rays arrive at the camera sensor and a clean scene is consequently mixed up with various opaque and semi-transparent artifacts. Existing deep learning methods are always constrained with limited real image pairs for training. Though recent synthesis-based approaches are found effective, synthesized pairs still deviate from the real ones as the mixing mechanism of flare artifacts and scenes in the wild always depends on a line of undetermined factors, such as lens structure, scratches, etc. In this paper, we present a new perspective from the blind nature of the flare removal task in a knowledge-driven manner. Specifically, we present a simple yet effective flare-level estimator to predict the corruption level of a flare-corrupted image. The estimated flare-level can be interpreted as additive information of the gap between corrupted images and their flare-free correspondences to facilitate a network at both training and testing stages adaptively. Besides, we utilize a flare-level modulator to better integrate the estimations into networks. We also devise a flare-aware block for more accurate flare recognition and reconstruction. Additionally, we collect a new real-world flare dataset for benchmarking, namely WiderFlare. Extensive experiments on three benchmark datasets demonstrate that our method outperforms state-of-the-art methods quantitatively and qualitatively.
{"title":"Toward Blind Flare Removal Using Knowledge-Driven Flare-Level Estimator","authors":"Haoyou Deng;Lida Li;Feng Zhang;Zhiqiang Li;Bin Xu;Qingbo Lu;Changxin Gao;Nong Sang","doi":"10.1109/TIP.2024.3480696","DOIUrl":"10.1109/TIP.2024.3480696","url":null,"abstract":"Lens flare is a common phenomenon when strong light rays arrive at the camera sensor and a clean scene is consequently mixed up with various opaque and semi-transparent artifacts. Existing deep learning methods are always constrained with limited real image pairs for training. Though recent synthesis-based approaches are found effective, synthesized pairs still deviate from the real ones as the mixing mechanism of flare artifacts and scenes in the wild always depends on a line of undetermined factors, such as lens structure, scratches, etc. In this paper, we present a new perspective from the blind nature of the flare removal task in a knowledge-driven manner. Specifically, we present a simple yet effective flare-level estimator to predict the corruption level of a flare-corrupted image. The estimated flare-level can be interpreted as additive information of the gap between corrupted images and their flare-free correspondences to facilitate a network at both training and testing stages adaptively. Besides, we utilize a flare-level modulator to better integrate the estimations into networks. We also devise a flare-aware block for more accurate flare recognition and reconstruction. Additionally, we collect a new real-world flare dataset for benchmarking, namely WiderFlare. Extensive experiments on three benchmark datasets demonstrate that our method outperforms state-of-the-art methods quantitatively and qualitatively.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6114-6128"},"PeriodicalIF":0.0,"publicationDate":"2024-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142486818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the increasing availability of multi-view data, multi-view representation learning has emerged as a prominent research area. However, collecting strictly view-aligned data is usually expensive, and learning from both aligned and unaligned data can be more practicable. Therefore, Partially View-aligned Representation Learning (PVRL) has recently attracted increasing attention. After aligning multi-view representations based on their semantic similarity, the aligned representations can be utilized to facilitate downstream tasks, such as clustering. However, existing methods may be constrained by the following limitations: 1) They learn semantic relations across views using the known correspondences, which is incomplete and the existence of false negative pairs (FNP) can significantly impact the learning effectiveness; 2) Existing strategies for alleviating the impact of FNP are too intuitive and lack a theoretical explanation of their applicable conditions; 3) They attempt to find FNP based on distance in the common space and fail to explore semantic relations between multi-view data. In this paper, we propose a Nonparametric Clustering-guided Cross-view Contrastive Learning (NC3L) for PVRL, in order to address the above issues. Firstly, we propose to estimate the similarity matrix between multi-view data in the marginal cross-view contrastive loss to approximate the similarity matrix of supervised contrastive learning (CL). Secondly, we establish the theoretical foundation for our proposed method by analyzing the error bounds of the loss function and its derivatives between our method and supervised CL. Thirdly, we propose a Deep Variational Nonparametric Clustering (DeepVNC) by designing a deep reparameterized variational inference for Dirichlet process Gaussian mixture models to construct cluster-level similarity between multi-view data and discover FNP. Additionally, we propose a reparameterization trick to improve the robustness and the performance of our proposed CL method. Extensive experiments on four widely used benchmark datasets show the superiority of our proposed method compared with state-of-the-art methods.
{"title":"Nonparametric Clustering-Guided Cross-View Contrastive Learning for Partially View-Aligned Representation Learning","authors":"Shengsheng Qian;Dizhan Xue;Jun Hu;Huaiwen Zhang;Changsheng Xu","doi":"10.1109/TIP.2024.3480701","DOIUrl":"10.1109/TIP.2024.3480701","url":null,"abstract":"With the increasing availability of multi-view data, multi-view representation learning has emerged as a prominent research area. However, collecting strictly view-aligned data is usually expensive, and learning from both aligned and unaligned data can be more practicable. Therefore, Partially View-aligned Representation Learning (PVRL) has recently attracted increasing attention. After aligning multi-view representations based on their semantic similarity, the aligned representations can be utilized to facilitate downstream tasks, such as clustering. However, existing methods may be constrained by the following limitations: 1) They learn semantic relations across views using the known correspondences, which is incomplete and the existence of false negative pairs (FNP) can significantly impact the learning effectiveness; 2) Existing strategies for alleviating the impact of FNP are too intuitive and lack a theoretical explanation of their applicable conditions; 3) They attempt to find FNP based on distance in the common space and fail to explore semantic relations between multi-view data. In this paper, we propose a Nonparametric Clustering-guided Cross-view Contrastive Learning (NC3L) for PVRL, in order to address the above issues. Firstly, we propose to estimate the similarity matrix between multi-view data in the marginal cross-view contrastive loss to approximate the similarity matrix of supervised contrastive learning (CL). Secondly, we establish the theoretical foundation for our proposed method by analyzing the error bounds of the loss function and its derivatives between our method and supervised CL. Thirdly, we propose a Deep Variational Nonparametric Clustering (DeepVNC) by designing a deep reparameterized variational inference for Dirichlet process Gaussian mixture models to construct cluster-level similarity between multi-view data and discover FNP. Additionally, we propose a reparameterization trick to improve the robustness and the performance of our proposed CL method. Extensive experiments on four widely used benchmark datasets show the superiority of our proposed method compared with state-of-the-art methods.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6158-6172"},"PeriodicalIF":0.0,"publicationDate":"2024-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142486817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-18DOI: 10.1109/TIP.2024.3477350
Gang Wu;Junjun Jiang;Junpeng Jiang;Xianming Liu
Recent progress in single-image super-resolution (SISR) has achieved remarkable performance, yet the computational costs of these methods remain a challenge for deployment on resource-constrained devices. In particular, transformer-based methods, which leverage self-attention mechanisms, have led to significant breakthroughs but also introduce substantial computational costs. To tackle this issue, we introduce the Convolutional Transformer layer (ConvFormer) and propose a ConvFormer-based Super-Resolution network (CFSR), offering an effective and efficient solution for lightweight image super-resolution. The proposed method inherits the advantages of both convolution-based and transformer-based approaches. Specifically, CFSR utilizes large kernel convolutions as a feature mixer to replace the self-attention module, efficiently modeling long-range dependencies and extensive receptive fields with minimal computational overhead. Furthermore, we propose an edge-preserving feed-forward network (EFN) designed to achieve local feature aggregation while effectively preserving high-frequency information. Extensive experiments demonstrate that CFSR strikes an optimal balance between computational cost and performance compared to existing lightweight SR methods. When benchmarked against state-of-the-art methods such as ShuffleMixer, the proposed CFSR achieves a gain of 0.39 dB on the Urban100 dataset for the x2 super-resolution task while requiring 26% and 31% fewer parameters and FLOPs, respectively. The code and pre-trained models are available at https://github.com/Aitical/CFSR .
{"title":"Transforming Image Super-Resolution: A ConvFormer-Based Efficient Approach","authors":"Gang Wu;Junjun Jiang;Junpeng Jiang;Xianming Liu","doi":"10.1109/TIP.2024.3477350","DOIUrl":"10.1109/TIP.2024.3477350","url":null,"abstract":"Recent progress in single-image super-resolution (SISR) has achieved remarkable performance, yet the computational costs of these methods remain a challenge for deployment on resource-constrained devices. In particular, transformer-based methods, which leverage self-attention mechanisms, have led to significant breakthroughs but also introduce substantial computational costs. To tackle this issue, we introduce the Convolutional Transformer layer (ConvFormer) and propose a ConvFormer-based Super-Resolution network (CFSR), offering an effective and efficient solution for lightweight image super-resolution. The proposed method inherits the advantages of both convolution-based and transformer-based approaches. Specifically, CFSR utilizes large kernel convolutions as a feature mixer to replace the self-attention module, efficiently modeling long-range dependencies and extensive receptive fields with minimal computational overhead. Furthermore, we propose an edge-preserving feed-forward network (EFN) designed to achieve local feature aggregation while effectively preserving high-frequency information. Extensive experiments demonstrate that CFSR strikes an optimal balance between computational cost and performance compared to existing lightweight SR methods. When benchmarked against state-of-the-art methods such as ShuffleMixer, the proposed CFSR achieves a gain of 0.39 dB on the Urban100 dataset for the x2 super-resolution task while requiring 26% and 31% fewer parameters and FLOPs, respectively. The code and pre-trained models are available at \u0000<uri>https://github.com/Aitical/CFSR</uri>\u0000.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6071-6082"},"PeriodicalIF":0.0,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142449611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-17DOI: 10.1109/TIP.2024.3478854
Jingren Liu;Zhong Ji;Yanwei Pang;Yunlong Yu
The proliferation of Few-Shot Class Incremental Learning (FSCIL) methodologies has highlighted the critical challenge of maintaining robust anti-amnesia capabilities in FSCIL learners. In this paper, we present a novel conceptualization of anti-amnesia in terms of mathematical generalization, leveraging the Neural Tangent Kernel (NTK) perspective. Our method focuses on two key aspects: ensuring optimal NTK convergence and minimizing NTK-related generalization loss, which serve as the theoretical foundation for cross-task generalization. To achieve global NTK convergence, we introduce a principled meta-learning mechanism that guides optimization within an expanded network architecture. Concurrently, to reduce the NTK-related generalization loss, we systematically optimize its constituent factors. Specifically, we initiate self-supervised pre-training on the base session to enhance NTK-related generalization potential. These self-supervised weights are then carefully refined through curricular alignment, followed by the application of dual NTK regularization tailored specifically for both convolutional and linear layers. Through the combined effects of these measures, our network acquires robust NTK properties, ensuring optimal convergence and stability of the NTK matrix and minimizing the NTK-related generalization loss, significantly enhancing its theoretical generalization. On popular FSCIL benchmark datasets, our NTK-FSCIL surpasses contemporary state-of-the-art approaches, elevating end-session accuracy by 2.9% to 9.3%.
{"title":"NTK-Guided Few-Shot Class Incremental Learning","authors":"Jingren Liu;Zhong Ji;Yanwei Pang;Yunlong Yu","doi":"10.1109/TIP.2024.3478854","DOIUrl":"10.1109/TIP.2024.3478854","url":null,"abstract":"The proliferation of Few-Shot Class Incremental Learning (FSCIL) methodologies has highlighted the critical challenge of maintaining robust anti-amnesia capabilities in FSCIL learners. In this paper, we present a novel conceptualization of anti-amnesia in terms of mathematical generalization, leveraging the Neural Tangent Kernel (NTK) perspective. Our method focuses on two key aspects: ensuring optimal NTK convergence and minimizing NTK-related generalization loss, which serve as the theoretical foundation for cross-task generalization. To achieve global NTK convergence, we introduce a principled meta-learning mechanism that guides optimization within an expanded network architecture. Concurrently, to reduce the NTK-related generalization loss, we systematically optimize its constituent factors. Specifically, we initiate self-supervised pre-training on the base session to enhance NTK-related generalization potential. These self-supervised weights are then carefully refined through curricular alignment, followed by the application of dual NTK regularization tailored specifically for both convolutional and linear layers. Through the combined effects of these measures, our network acquires robust NTK properties, ensuring optimal convergence and stability of the NTK matrix and minimizing the NTK-related generalization loss, significantly enhancing its theoretical generalization. On popular FSCIL benchmark datasets, our NTK-FSCIL surpasses contemporary state-of-the-art approaches, elevating end-session accuracy by 2.9% to 9.3%.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"6029-6044"},"PeriodicalIF":0.0,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142448771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-17DOI: 10.1109/TIP.2024.3477356
Mu Li;Youneng Bao;Xiaohang Sui;Jinxing Li;Guangming Lu;Yong Xu
Learned 360° image compression methods using equirectangular projection (ERP) often confront a non-uniform sampling issue, inherent to sphere-to-rectangle projection. While uniformly or nearly uniformly sampling representations, along with their corresponding convolution operations, have been proposed to mitigate this issue, these methods often concentrate solely on uniform sampling rates, thus neglecting the content of the image. In this paper, we urge that different contents within 360° images have varying significance and advocate for the adoption of a content-adaptive parametric representation in 360° image compression, which takes into account both the content and sampling rate. We first introduce the parametric pseudocylindrical representation and corresponding convolution operation, upon which we build a learned 360° image codec. Then, we model the hyperparameter of the representation as the output of a network, derived from the image’s content and its spherical coordinates. We treat the optimization of hyperparameters for different 360° images as distinct compression tasks and propose a meta-learning algorithm to jointly optimize the codec and the metaknowledge, i.e., the hyperparameter estimation network. A significant challenge is the lack of a direct derivative from the compression loss to the hyperparameter network. To address this, we present a novel method to relax the rate-distortion loss as a function of the hyperparameters, enabling gradient-based optimization of the metaknowledge. Experimental results on omnidirectional images demonstrate that our method achieves state-of-the-art performance and superior visual quality.
{"title":"Learning Content-Weighted Pseudocylindrical Representation for 360° Image Compression","authors":"Mu Li;Youneng Bao;Xiaohang Sui;Jinxing Li;Guangming Lu;Yong Xu","doi":"10.1109/TIP.2024.3477356","DOIUrl":"10.1109/TIP.2024.3477356","url":null,"abstract":"Learned 360° image compression methods using equirectangular projection (ERP) often confront a non-uniform sampling issue, inherent to sphere-to-rectangle projection. While uniformly or nearly uniformly sampling representations, along with their corresponding convolution operations, have been proposed to mitigate this issue, these methods often concentrate solely on uniform sampling rates, thus neglecting the content of the image. In this paper, we urge that different contents within 360° images have varying significance and advocate for the adoption of a content-adaptive parametric representation in 360° image compression, which takes into account both the content and sampling rate. We first introduce the parametric pseudocylindrical representation and corresponding convolution operation, upon which we build a learned 360° image codec. Then, we model the hyperparameter of the representation as the output of a network, derived from the image’s content and its spherical coordinates. We treat the optimization of hyperparameters for different 360° images as distinct compression tasks and propose a meta-learning algorithm to jointly optimize the codec and the metaknowledge, i.e., the hyperparameter estimation network. A significant challenge is the lack of a direct derivative from the compression loss to the hyperparameter network. To address this, we present a novel method to relax the rate-distortion loss as a function of the hyperparameters, enabling gradient-based optimization of the metaknowledge. Experimental results on omnidirectional images demonstrate that our method achieves state-of-the-art performance and superior visual quality.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"33 ","pages":"5975-5988"},"PeriodicalIF":0.0,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142448772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：