We introduce a novel energy formulation for Plug- and-Play (PnP) image recovery. Traditional PnP methods that use a convolutional neural network (CNN) do not have an energy based formulation. The primary focus of this work is to introduce an energy-based PnP formulation, which relies on a CNN that learns the log of the image prior from training data. The score function is evaluated as the gradient of the energy model, which resembles a UNET with shared encoder and decoder weights. The proposed score function is thus constrained to a conservative vector field, which is the key difference with classical PnP models. The energy-based formulation offers algorithms with convergence guarantees, even when the learned score model is not a contraction. The relaxation of the contraction constraint allows the proposed model to learn more complex priors, thus offering improved performance over traditional PnP schemes. Our experiments in magnetic resonance image reconstruction demonstrates the improved performance offered by the proposed energy model over traditional PnP methods.
Cracks play a crucial role in assessing the safety and durability of manufactured buildings. However, the long and sharp topological features and complex background of cracks make the task of crack segmentation extremely challenging. In this paper, we propose a novel convolutional-transformer network based on encoder-decoder architecture to solve this challenge. Particularly, we designed a Dilated Residual Block (DRB) and a Boundary Awareness Module (BAM). The DRB pays attention to the local detail of cracks and adjusts the feature dimension for other blocks as needed. And the BAM learns the boundary features from the dilated crack label. Furthermore, the DRB is combined with a lightweight transformer that captures global information to serve as an effective encoder. Experimental results show that the proposed network performs better than state-of-the-art algorithms on two typical datasets. Datasets, code, and trained models are available for research at https://github.com/HqiTao/CT-crackseg.
Denoising Diffusion models have shown remarkable performance in generating diverse, high quality images from text. Numerous techniques have been proposed on top of or in alignment with models like Stable Diffusion and Imagen that generate images directly from text. A lesser explored approach is DALLE-2's two step process comprising a Diffusion Prior that generates a CLIP image embedding from text and a Diffusion Decoder that generates an image from a CLIP image embedding. We explore the capabilities of the Diffusion Prior and the advantages of an intermediate CLIP representation. We observe that Diffusion Prior can be used in a memory and compute efficient way to constrain the generation to a specific domain without altering the larger Diffusion Decoder. Moreover, we show that the Diffusion Prior can be trained with additional conditional information such as color histogram to further control the generation. We show quantitatively and qualitatively that the proposed approaches perform better than prompt engineering for domain specific generation and existing baselines for color conditioned generation. We believe that our observations and results will instigate further research into the diffusion prior and uncover more of its capabilities.
The recovery of time-varying graph signals is a fundamental problem with numerous applications in sensor networks and forecasting in time series. Effectively capturing the spatio-temporal information in these signals is essential for the downstream tasks. Previous studies have used the smoothness of the temporal differences of such graph signals as an initial assumption. Nevertheless, this smoothness assumption could result in a degradation of performance in the corresponding application when the prior does not hold. In this work, we relax the requirement of this hypothesis by including a learning module. We propose a Time Graph Neural Network (TimeGNN) for the recovery of time-varying graph signals. Our algorithm uses an encoder-decoder architecture with a specialized loss composed of a mean squared error function and a Sobolev smoothness operator.TimeGNN shows competitive performance against previous methods in real datasets.
Human motion transfer aims to transfer motions from a target dynamic person to a source static one for motion synthesis. An accurate matching between the source person and the target motion in both large and subtle motion changes is vital for improving the transferred motion quality. In this paper, we propose Human MotionFormer, a hierarchical ViT framework that leverages global and local perceptions to capture large and subtle motion matching, respectively. It consists of two ViT encoders to extract input features (i.e., a target motion image and a source human image) and a ViT decoder with several cascaded blocks for feature matching and motion transfer. In each block, we set the target motion feature as Query and the source person as Key and Value, calculating the cross-attention maps to conduct a global feature matching. Further, we introduce a convolutional layer to improve the local perception after the global cross-attention computations. This matching process is implemented in both warping and generation branches to guide the motion transfer. During training, we propose a mutual learning loss to enable the co-supervision between warping and generation branches for better motion representations. Experiments show that our Human MotionFormer sets the new state-of-the-art performance both qualitatively and quantitatively. Project page: \url{https://github.com/KumapowerLIU/Human-MotionFormer}
This paper presents a deep learning framework for medical video segmentation. Convolution neural network (CNN) and transformer-based methods have achieved great milestones in medical image segmentation tasks due to their incredible semantic feature encoding and global information comprehension abilities. However, most existing approaches ignore a salient aspect of medical video data - the temporal dimension. Our proposed framework explicitly extracts features from neighbouring frames across the temporal dimension and incorporates them with a temporal feature blender, which then tokenises the high-level spatio-temporal feature to form a strong global feature encoded via a Swin Transformer. The final segmentation results are produced via a UNet-like encoder-decoder architecture. Our model outperforms other approaches by a significant margin and improves the segmentation benchmarks on the VFSS2022 dataset, achieving a dice coefficient of 0.8986 and 0.8186 for the two datasets tested. Our studies also show the efficacy of the temporal feature blending scheme and cross-dataset transferability of learned capabilities. Code and models are fully available at https://github.com/SimonZeng7108/Video-SwinUNet.
For real-time speech enhancement (SE) including noise suppression, dereverberation and acoustic echo cancellation, the time-variance of the audio signals becomes a severe challenge. The causality and memory usage limit that only the historical information can be used for the system to capture the time-variant characteristics. We propose to adaptively change the receptive field according to the input signal in deep neural network based SE model. Specifically, in an encoder-decoder framework, a dynamic attention span mechanism is introduced to all the attention modules for controlling the size of historical content used for processing the current frame. Experimental results verify that this dynamic mechanism can better track time-variant factors and capture speech-related characteristics, benefiting to both interference removing and speech quality retaining.
Deep learning models have been widely applied for fast MRI. The majority of existing deep learning models, e.g., convolutional neural networks, work on data with Euclidean or regular grids structures. However, high-dimensional features extracted from MR data could be encapsulated in non-Euclidean manifolds. This disparity between the go-to assumption of existing models and data requirements limits the flexibility to capture irregular anatomical features in MR data. In this work, we introduce a novel Vision GNN type network for fast MRI called Vision GNN U-Net (ViGU). More precisely, the pixel array is first embedded into patches and then converted into a graph. Secondly, a U-shape network is developed using several graph blocks in symmetrical encoder and decoder paths. Moreover, we show that the proposed ViGU can also benefit from Generative Adversarial Networks yielding to its variant ViGU-GAN. We demonstrate, through numerical and visual experiments, that the proposed ViGU and GAN variant outperform existing CNN and GAN-based methods. Moreover, we show that the proposed network readily competes with approaches based on Transformers while requiring a fraction of the computational cost. More importantly, the graph structure of the network reveals how the network extracts features from MR images, providing intuitive explainability.
Research in the field of Continual Semantic Segmentation is mainly investigating novel learning algorithms to overcome catastrophic forgetting of neural networks. Most recent publications have focused on improving learning algorithms without distinguishing effects caused by the choice of neural architecture.Therefore, we study how the choice of neural network architecture affects catastrophic forgetting in class- and domain-incremental semantic segmentation. Specifically, we compare the well-researched CNNs to recently proposed Transformers and Hybrid architectures, as well as the impact of the choice of novel normalization layers and different decoder heads. We find that traditional CNNs like ResNet have high plasticity but low stability, while transformer architectures are much more stable. When the inductive biases of CNN architectures are combined with transformers in hybrid architectures, it leads to higher plasticity and stability. The stability of these models can be explained by their ability to learn general features that are robust against distribution shifts. Experiments with different normalization layers show that Continual Normalization achieves the best trade-off in terms of adaptability and stability of the model. In the class-incremental setting, the choice of the normalization layer has much less impact. Our experiments suggest that the right choice of architecture can significantly reduce forgetting even with naive fine-tuning and confirm that for real-world applications, the architecture is an important factor in designing a continual learning model.
Existing studies tend tofocus onmodel modifications and integration with higher accuracy, which improve performance but also carry huge computational costs, resulting in longer detection times. Inmedical imaging, the use of time is extremely sensitive. And at present most of the semantic segmentation models have encoder-decoder structure or double branch structure. Their several times of the pooling use with high-level semantic information extraction operation cause information loss although there si a reverse pooling or other similar action to restore information loss of pooling operation. In addition, we notice that visual attention mechanism has superior performance on a variety of tasks. Given this, this paper proposes non-pooling network(NPNet), non-pooling commendably reduces the loss of information and attention enhancement m o d u l e ( A M ) effectively increases the weight of useful information. The method greatly reduces the number of parametersand computation costs by the shallow neural network structure. We evaluate the semantic segmentation model of our NPNet on three benchmark datasets comparing w i t h multiple current state-of-the-art(SOTA) models, and the implementation results show thatour NPNetachieves SOTA performance, with an excellent balance between accuracyand speed.
Costly, noisy, and over-specialized, labels are to be set aside in favor of unsupervised learning if we hope to learn cheap, reliable, and transferable models. To that end, spectral embedding, self-supervised learning, or generative modeling have offered competitive solutions. Those methods however come with numerous challenges \textit{e.g.} estimating geodesic distances, specifying projector architectures and anti-collapse losses, or specifying decoder architectures and reconstruction losses. In contrast, we introduce a simple explainable alternative -- coined \textbf{DIET} -- to learn representations from unlabeled data, free of those challenges. \textbf{DIET} is blatantly simple: take one's favorite classification setup and use the \textbf{D}atum \textbf{I}nd\textbf{E}x as its \textbf{T}arget class, \textit{i.e. each sample is its own class}, no further changes needed. \textbf{DIET} works without a decoder/projector network, is not based on positive pairs nor reconstruction, introduces no hyper-parameters, and works out-of-the-box across datasets and architectures. Despite \textbf{DIET}'s simplicity, the learned representations are of high-quality and often on-par with the state-of-the-art \textit{e.g.} using a linear classifier on top of DIET's learned representation reaches $71.4\%$ on CIFAR100 with a Resnet101, $52.5\%$ on TinyImagenet with a Resnext50.
In this paper, we propose an iterative source error correction (ISEC) decoding scheme for deep-learning-based joint source-channel coding (Deep JSCC). Given a noisy codeword received through the channel, we use a Deep JSCC encoder and decoder pair to update the codeword iteratively to find a (modified) maximum a-posteriori (MAP) solution. For efficient MAP decoding, we utilize a neural network-based denoiser to approximate the gradient of the log-prior density of the codeword space. Albeit the non-convexity of the optimization problem, our proposed scheme improves various distortion and perceptual quality metrics from the conventional one-shot (non-iterative) Deep JSCC decoding baseline. Furthermore, the proposed scheme produces more reliable source reconstruction results compared to the baseline when the channel noise characteristics do not match the ones used during training.
Image reconstruction-based anomaly detection has recently been in the spotlight because of the difficulty of constructing anomaly datasets. These approaches work by learning to model normal features without seeing abnormal samples during training and then discriminating anomalies at test time based on the reconstructive errors. However, these models have limitations in reconstructing the abnormal samples due to their indiscriminate conveyance of features. Moreover, these approaches are not explicitly optimized for distinguishable anomalies. To address these problems, we propose a two-stream decoder network (TSDN), designed to learn both normal and abnormal features. Additionally, we propose a feature normality estimator (FNE) to eliminate abnormal features and prevent high-quality reconstruction of abnormal regions. Evaluation on a standard benchmark demonstrated performance better than state-of-the-art models.
Accurate visualization of liver tumors and their surrounding blood vessels is essential for noninvasive diagnosis and prognosis prediction of tumors. In medical image segmentation, there is still a lack of in-depth research on the simultaneous segmentation of liver tumors and peritumoral blood vessels. To this end, we collect the first liver tumor, and vessel segmentation benchmark datasets containing 52 portal vein phase computed tomography images with liver, liver tumor, and vessel annotations. In this case, we propose a 3D U-shaped Cross-Attention Network (UCA-Net) that utilizes a tailored cross-attention mechanism instead of the traditional skip connection to effectively model the encoder and decoder feature. Specifically, the UCA-Net uses a channel-wise cross-attention module to reduce the semantic gap between encoder and decoder and a slice-wise cross-attention module to enhance the contextual semantic learning ability among distinct slices. Experimental results show that the proposed UCA-Net can accurately segment 3D medical images and achieve state-of-the-art performance on the liver tumor and intrahepatic vessel segmentation task.
Image restoration is a long-standing low-level vision problem, e.g., deblurring and deraining. In the process of image restoration, it is necessary to consider not only the spatial details and contextual information of restoration to ensure the quality, but also the system complexity. Although many methods have been able to guarantee the quality of image restoration, the system complexity of the state-of-the-art (SOTA) methods is increasing as well. Motivated by this, we present a mixed hierarchy network that can balance these competing goals. Our main proposal is a mixed hierarchy architecture, that progressively recovers contextual information and spatial details from degraded images while we design intra-blocks to reduce system complexity. Specifically, our model first learns the contextual information using encoder-decoder architectures, and then combines them with high-resolution branches that preserve spatial detail. In order to reduce the system complexity of this architecture for convenient analysis and comparison, we replace or remove the nonlinear activation function with multiplication and use a simple network structure. In addition, we replace spatial convolution with global self-attention for the middle block of encoder-decoder. The resulting tightly interlinked hierarchy architecture, named as MHNet, delivers strong performance gains on several image restoration tasks, including image deraining, and deblurring.
Large pre-trained language models (PLMs) have been shown to retain implicit knowledge within their parameters. To enhance this implicit knowledge, we propose Knowledge Injection into Language Models (KILM), a novel approach that injects entity-related knowledge into encoder-decoder PLMs, via a generative knowledge infilling objective through continued pre-training. This is done without architectural modifications to the PLMs or adding additional parameters. Experimental results over a suite of knowledge-intensive tasks spanning numerous datasets show that KILM enables models to retain more knowledge and hallucinate less, while preserving their original performance on general NLU and NLG tasks. KILM also demonstrates improved zero-shot performances on tasks such as entity disambiguation, outperforming state-of-the-art models having 30x more parameters.
VAEs are probabilistic graphical models based on neural networks that allow the coding of input data in a latent space formed by simpler probability distributions and the reconstruction, based on such latent variables, of the source data. After training, the reconstruction network, called decoder, is capable of generating new elements belonging to a close distribution, ideally equal to the original one. This article has been written in Spanish to facilitate the arrival of this scientific knowledge to the Spanish-speaking community.
Exploring the complementary information of multi-view data to improve clustering effects is a crucial issue in multi-view clustering. In this paper, we propose a novel model based on information theory termed Informative Multi-View Clustering (IMVC), which extracts the common and view-specific information hidden in multi-view data and constructs a clustering-oriented comprehensive representation. More specifically, we concatenate multiple features into a unified feature representation, then pass it through a encoder to retrieve the common representation across views. Simultaneously, the features of each view are sent to a encoder to produce a compact view-specific representation, respectively. Thus, we constrain the mutual information between the common representation and view-specific representations to be minimal for obtaining multi-level information. Further, the common representation and view-specific representation are spliced to model the refined representation of each view, which is fed into a decoder to reconstruct the initial data with maximizing their mutual information. In order to form a comprehensive representation, the common representation and all view-specific representations are concatenated. Furthermore, to accommodate the comprehensive representation better for the clustering task, we maximize the mutual information between an instance and its k-nearest neighbors to enhance the intra-cluster aggregation, thus inducing well separation of different clusters at the overall aspect. Finally, we conduct extensive experiments on six benchmark datasets, and the experimental results indicate that the proposed IMVC outperforms other methods.
This paper proposes to use both audio input and subject information to predict the personalized preference of two audio segments with the same content in different qualities. A siamese network is used to compare the inputs and predict the preference. Several different structures for each side of the siamese network are investigated, and an LDNet with PANNs' CNN6 as the encoder and a multi-layer perceptron block as the decoder outperforms a baseline model using only audio input the most, where the overall accuracy grows from 77.56% to 78.04%. Experimental results also show that using all the subject information, including age, gender, and the specifications of headphones or earphones, is more effective than using only a part of them.
Diffusion models have shown remarkable capabilities in generating high quality and creative images conditioned on text. An interesting application of such models is structure preserving text guided image editing. Existing approaches rely on text conditioned diffusion models such as Stable Diffusion or Imagen and require compute intensive optimization of text embeddings or fine-tuning the model weights for text guided image editing. We explore text guided image editing with a Hybrid Diffusion Model (HDM) architecture similar to DALLE-2. Our architecture consists of a diffusion prior model that generates CLIP image embedding conditioned on a text prompt and a custom Latent Diffusion Model trained to generate images conditioned on CLIP image embedding. We discover that the diffusion prior model can be used to perform text guided conceptual edits on the CLIP image embedding space without any finetuning or optimization. We combine this with structure preserving edits on the image decoder using existing approaches such as reverse DDIM to perform text guided image editing. Our approach, PRedItOR does not require additional inputs, fine-tuning, optimization or objectives and shows on par or better results than baselines qualitatively and quantitatively. We provide further analysis and understanding of the diffusion prior model and believe this opens up new possibilities in diffusion models research.