PyTorchKR Papers

최신 논문

총 9,642편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2201.10953

Dual-Tasks Siamese Transformer Framework for Building Damage Assessment

Accurate and fine-grained information about the extent of damage to buildings is essential for humanitarian relief and disaster response. However, as the most commonly used architecture in remote sensing interpretation tasks, Convolutional Neural Networks (CNNs) have limited ability to model the non-local relationship between pixels. Recently, Transformer architecture first proposed for modeling long-range dependency in natural language processing has shown promising results in computer vision tasks. Considering the frontier advances of Transformer architecture in the computer vision field, in this paper, we present the first attempt at designing a Transformer-based damage assessment architecture (DamFormer). In DamFormer, a siamese Transformer encoder is first constructed to extract non-local and representative deep features from input multitemporal image-pairs. Then, a multitemporal fusion module is designed to fuse information for downstream tasks. Finally, a lightweight dual-tasks decoder aggregates multi-level features for final prediction. To the best of our knowledge, it is the first time that such a deep Transformer-based network is proposed for multitemporal remote sensing interpretation tasks. The experimental results on the large-scale damage assessment dataset xBD demonstrate the potential of the Transformer-based architecture.

arXiv:2201.10753공식 코드 · Papers with Code

Interactive Image Inpainting Using Semantic Guidance

Image inpainting approaches have achieved significant progress with the help of deep neural networks. However, existing approaches mainly focus on leveraging the priori distribution learned by neural networks to produce a single inpainting result or further yielding multiple solutions, where the controllability is not well studied. This paper develops a novel image inpainting approach that enables users to customize the inpainting result by their own preference or memory. Specifically, our approach is composed of two stages that utilize the prior of neural network and user's guidance to jointly inpaint corrupted images. In the first stage, an autoencoder based on a novel external spatial attention mechanism is deployed to produce reconstructed features of the corrupted image and a coarse inpainting result that provides semantic mask as the medium for user interaction. In the second stage, a semantic decoder that takes the reconstructed features as prior is adopted to synthesize a fine inpainting result guided by user's customized semantic mask, so that the final inpainting result will share the same content with user's guidance while the textures and colors reconstructed in the first stage are preserved. Extensive experiments demonstrate the superiority of our approach in terms of inpainting quality and controllability.

arXiv:2201.10736

A Joint Convolution Auto-encoder Network for Infrared and Visible Image Fusion

Background: Leaning redundant and complementary relationships is a critical step in the human visual system. Inspired by the infrared cognition ability of crotalinae animals, we design a joint convolution auto-encoder (JCAE) network for infrared and visible image fusion. Methods: Our key insight is to feed infrared and visible pair images into the network simultaneously and separate an encoder stream into two private branches and one common branch, the private branch works for complementary features learning and the common branch does for redundant features learning. We also build two fusion rules to integrate redundant and complementary features into their fused feature which are then fed into the decoder layer to produce the final fused image. We detail the structure, fusion rule and explain its multi-task loss function. Results: Our JCAE network achieves good results in terms of both subjective effect and objective evaluation metrics.

arXiv:2201.10703공식 코드 · Papers with Code

Anomaly Detection via Reverse Distillation from One-Class Embedding

Knowledge distillation (KD) achieves promising results on the challenging problem of unsupervised anomaly detection (AD).The representation discrepancy of anomalies in the teacher-student (T-S) model provides essential evidence for AD. However, using similar or identical architectures to build the teacher and student models in previous studies hinders the diversity of anomalous representations. To tackle this problem, we propose a novel T-S model consisting of a teacher encoder and a student decoder and introduce a simple yet effective "reverse distillation" paradigm accordingly. Instead of receiving raw images directly, the student network takes teacher model's one-class embedding as input and targets to restore the teacher's multiscale representations. Inherently, knowledge distillation in this study starts from abstract, high-level presentations to low-level features. In addition, we introduce a trainable one-class bottleneck embedding (OCBE) module in our T-S model. The obtained compact embedding effectively preserves essential information on normal patterns, but abandons anomaly perturbations. Extensive experimentation on AD and one-class novelty detection benchmarks shows that our method surpasses SOTA performance, demonstrating our proposed approach's effectiveness and generalizability.

arXiv:2201.10649공식 코드 · Papers with Code

Attentive Task Interaction Network for Multi-Task Learning

Multitask learning (MTL) has recently gained a lot of popularity as a learning paradigm that can lead to improved per-task performance while also using fewer per-task model parameters compared to single task learning. One of the biggest challenges regarding MTL networks involves how to share features across tasks. To address this challenge, we propose the Attentive Task Interaction Network (ATI-Net). ATI-Net employs knowledge distillation of the latent features for each task, then combines the feature maps to provide improved contextualized information to the decoder. This novel approach to introducing knowledge distillation into an attention based multitask network outperforms state of the art MTL baselines such as the standalone MTAN and PAD-Net, with roughly the same number of model parameters.

arXiv:2201.10252공식 코드 · Papers with Code

DocEnTr: An End-to-End Document Image Enhancement Transformer

Document images can be affected by many degradation scenarios, which cause recognition and processing difficulties. In this age of digitization, it is important to denoise them for proper usage. To address this challenge, we present a new encoder-decoder architecture based on vision transformers to enhance both machine-printed and handwritten document images, in an end-to-end fashion. The encoder operates directly on the pixel patches with their positional information without the use of any convolutional layers, while the decoder reconstructs a clean image from the encoded patches. Conducted experiments show a superiority of the proposed model compared to the state-of the-art methods on several DIBCO benchmarks. Code and models will be publicly available at: \url{https://github.com/dali92002/DocEnTR}.

arXiv:2201.10190

Run-and-back stitch search: novel block synchronous decoding for streaming encoder-decoder ASR

A streaming style inference of encoder-decoder automatic speech recognition (ASR) system is important for reducing latency, which is essential for interactive use cases. To this end, we propose a novel blockwise synchronous decoding algorithm with a hybrid approach that combines endpoint prediction and endpoint post-determination. In the endpoint prediction, we compute the expectation of the number of tokens that are yet to be emitted in the encoder features of the current blocks using the CTC posterior. Based on the expectation value, the decoder predicts the endpoint to realize continuous block synchronization, as a running stitch. Meanwhile, endpoint post-determination probabilistically detects backward jump of the source-target attention, which is caused by the misprediction of endpoints. Then it resumes decoding by discarding those hypotheses, as back stitch. We combine these methods into a hybrid approach, namely run-and-back stitch search, which reduces the computational cost and latency. Evaluations of various ASR tasks show the efficiency of our proposed decoding algorithm, which achieves a latency reduction, for instance in the Librispeech test set from 1487 ms to 821 ms at the 90th percentile, while maintaining a high recognition accuracy.

arXiv:2201.10449

An adaptive closed-loop ECoG decoder for long-term and stable bimanual control of an exoskeleton by a tetraplegic

Brain-computer interfaces (BCIs) still face many challenges to step out of laboratories to be used in real-life applications. A key one persists in the high performance control of diverse effectors for complex tasks, using chronic and safe recorders. This control must be robust over time and of high decoding performance without continuous recalibration of the decoders. In the article, asynchronous control of an exoskeleton by a tetraplegic patient using a chronically implanted epidural electrocorticography (EpiCoG) implant is demonstrated. For this purpose, an adaptive online tensor-based decoder: the Recursive Exponentially Weighted Markov-Switching multi-Linear Model (REW-MSLM) was developed. We demonstrated over a period of 6 months the stability of the 8-dimensional alternative bimanual control of the exoskeleton and its virtual avatar using REW-MSLM without recalibration of the decoder.

arXiv:2201.10340

Distributed Image Transmission using Deep Joint Source-Channel Coding

We study the problem of deep joint source-channel coding (D-JSCC) for correlated image sources, where each source is transmitted through a noisy independent channel to the common receiver. In particular, we consider a pair of images captured by two cameras with probably overlapping fields of view transmitted over wireless channels and reconstructed in the center node. The challenging problem involves designing a practical code to utilize both source and channel correlations to improve transmission efficiency without additional transmission overhead. To tackle this, we need to consider the common information across two stereo images as well as the differences between two transmission channels. In this case, we propose a deep neural networks solution that includes lightweight edge encoders and a powerful center decoder. Besides, in the decoder, we propose a novel channel state information aware cross attention module to highlight the overlapping fields and leverage the relevance between two noisy feature maps.Our results show the impressive improvement of reconstruction quality in both links by exploiting the noisy representations of the other link. Moreover, the proposed scheme shows competitive results compared to the separated schemes with capacity-achieving channel codes.

arXiv:2201.09834

Deep Decoding of $\ell_\infty$-coded Light Field Images

To enrich the functionalities of traditional cameras, light field cameras record both the intensity and direction of light rays, so that images can be rendered with user-defined camera parameters via computations. The added capability and flexibility are gained at the cost of gathering typically more than $100\times$ greater amount of information than conventional images. To cope with this issue, several light field compression schemes have been introduced. However, their ways of exploiting correlations of multidimensional light field data are complex and are hence not suited for inexpensive light field cameras. In this work, we propose a novel $\ell_\infty$-constrained light-field image compression system that has a very low-complexity DPCM encoder and a CNN-based deep decoder. Targeting high-fidelity reconstruction, the CNN decoder capitalizes on the $\ell_\infty$-constraint and light field properties to remove the compression artifacts and achieves significantly better performance than existing state-of-the-art $\ell_2$-based light field compression methods.

arXiv:2201.09429

End-to-End Neural Speech Coding for Real-Time Communications

Deep-learning based methods have shown their advantages in audio coding over traditional ones but limited attention has been paid on real-time communications (RTC). This paper proposes the TFNet, an end-to-end neural speech codec with low latency for RTC. It takes an encoder-temporal filtering-decoder paradigm that has seldom been investigated in audio coding. An interleaved structure is proposed for temporal filtering to capture both short-term and long-term temporal dependencies. Furthermore, with end-to-end optimization, the TFNet is jointly optimized with speech enhancement and packet loss concealment, yielding a one-for-all network for three tasks. Both subjective and objective results demonstrate the efficiency of the proposed TFNet.

arXiv:2201.09113

Predicting Physics in Mesh-reduced Space with Temporal Attention

Graph-based next-step prediction models have recently been very successful in modeling complex high-dimensional physical systems on irregular meshes. However, due to their short temporal attention span, these models suffer from error accumulation and drift. In this paper, we propose a new method that captures long-term dependencies through a transformer-style temporal attention model. We introduce an encoder-decoder structure to summarize features and create a compact mesh representation of the system state, to allow the temporal model to operate on a low-dimensional mesh representations in a memory efficient manner. Our method outperforms a competitive GNN baseline on several complex fluid dynamics prediction tasks, from sonic shocks to vascular flow. We demonstrate stable rollouts without the need for training noise and show perfectly phase-stable predictions even for very long sequences. More broadly, we believe our approach paves the way to bringing the benefits of attention-based sequence models to solving high-dimensional complex physics tasks.

arXiv:2201.08877

Variational Autoencoder based Metamodeling for Multi-Objective Topology Optimization of Electrical Machines

Conventional magneto-static finite element analysis of electrical machine design is time-consuming and computationally expensive. Since each machine topology has a distinct set of parameters, design optimization is commonly performed independently. This paper presents a novel method for predicting Key Performance Indicators (KPIs) of differently parameterized electrical machine topologies at the same time by mapping a high dimensional integrated design parameters in a lower dimensional latent space using a variational autoencoder. After training, via a latent space, the decoder and multi-layer neural network will function as meta-models for sampling new designs and predicting associated KPIs, respectively. This enables parameter-based concurrent multi-topology optimization.

arXiv:2201.09405공식 코드 · Papers with Code

Improving Chest X-Ray Report Generation by Leveraging Warm Starting

Automatically generating a report from a patient's Chest X-Rays (CXRs) is a promising solution to reducing clinical workload and improving patient care. However, current CXR report generators -- which are predominantly encoder-to-decoder models -- lack the diagnostic accuracy to be deployed in a clinical setting. To improve CXR report generation, we investigate warm starting the encoder and decoder with recent open-source computer vision and natural language processing checkpoints, such as the Vision Transformer (ViT) and PubMedBERT. To this end, each checkpoint is evaluated on the MIMIC-CXR and IU X-Ray datasets. Our experimental investigation demonstrates that the Convolutional vision Transformer (CvT) ImageNet-21K and the Distilled Generative Pre-trained Transformer 2 (DistilGPT2) checkpoints are best for warm starting the encoder and decoder, respectively. Compared to the state-of-the-art ($\mathcal{M}^2$ Transformer Progressive), CvT2DistilGPT2 attained an improvement of 8.3\% for CE F-1, 1.8\% for BLEU-4, 1.6\% for ROUGE-L, and 1.0\% for METEOR. The reports generated by CvT2DistilGPT2 have a higher similarity to radiologist reports than previous approaches. This indicates that leveraging warm starting improves CXR report generation. Code and checkpoints for CvT2DistilGPT2 are available at https://github.com/aehrc/cvt2distilgpt2.

arXiv:2201.09390공식 코드 · Papers with Code

AttentionHTR: Handwritten Text Recognition Based on Attention Encoder-Decoder Networks

This work proposes an attention-based sequence-to-sequence model for handwritten word recognition and explores transfer learning for data-efficient training of HTR systems. To overcome training data scarcity, this work leverages models pre-trained on scene text images as a starting point towards tailoring the handwriting recognition models. ResNet feature extraction and bidirectional LSTM-based sequence modeling stages together form an encoder. The prediction stage consists of a decoder and a content-based attention mechanism. The effectiveness of the proposed end-to-end HTR system has been empirically evaluated on a novel multi-writer dataset Imgur5K and the IAM dataset. The experimental results evaluate the performance of the HTR framework, further supported by an in-depth analysis of the error cases. Source code and pre-trained models are available at https://github.com/dmitrijsk/AttentionHTR.

arXiv:2201.08582공식 코드 · Papers with Code

SegTransVAE: Hybrid CNN -- Transformer with Regularization for medical image segmentation

Current research on deep learning for medical image segmentation exposes their limitations in learning either global semantic information or local contextual information. To tackle these issues, a novel network named SegTransVAE is proposed in this paper. SegTransVAE is built upon encoder-decoder architecture, exploiting transformer with the variational autoencoder (VAE) branch to the network to reconstruct the input images jointly with segmentation. To the best of our knowledge, this is the first method combining the success of CNN, transformer, and VAE. Evaluation on various recently introduced datasets shows that SegTransVAE outperforms previous methods in Dice Score and $95\%$-Haudorff Distance while having comparable inference time to a simple CNN-based architecture network. The source code is available at: https://github.com/itruonghai/SegTransVAE.

arXiv:2201.08264

End-to-end Generative Pretraining for Multimodal Video Captioning

Recent video and language pretraining frameworks lack the ability to generate sentences. We present Multimodal Video Generative Pretraining (MV-GPT), a new pretraining framework for learning from unlabelled videos which can be effectively used for generative tasks such as multimodal video captioning. Unlike recent video-language pretraining frameworks, our framework trains both a multimodal video encoder and a sentence decoder jointly. To overcome the lack of captions in unlabelled videos, we leverage the future utterance as an additional text source and propose a bidirectional generation objective -- we generate future utterances given the present mulitmodal context, and also the present utterance given future observations. With this objective, we train an encoder-decoder model end-to-end to generate a caption from raw pixels and transcribed speech directly. Our model achieves state-of-the-art performance for multimodal video captioning on four standard benchmarks, as well as for other video understanding tasks such as VideoQA, video retrieval and action classification.

arXiv:2201.08029

Domain Generalization via Frequency-domain-based Feature Disentanglement and Interaction

Adaptation to out-of-distribution data is a meta-challenge for all statistical learning algorithms that strongly rely on the i.i.d. assumption. It leads to unavoidable labor costs and confidence crises in realistic applications. For that, domain generalization aims at mining domain-irrelevant knowledge from multiple source domains that can generalize to unseen target domains. In this paper, by leveraging the frequency domain of an image, we uniquely work with two key observations: (i) the high-frequency information of an image depicts object edge structure, which preserves high-level semantic information of the object is naturally consistent across different domains, and (ii) the low-frequency component retains object smooth structure, while this information is susceptible to domain shifts. Motivated by the above observations, we introduce (i) an encoder-decoder structure to disentangle high- and low-frequency feature of an image, (ii) an information interaction mechanism to ensure the helpful knowledge from both two parts can cooperate effectively, and (iii) a novel data augmentation technique that works on the frequency domain to encourage the robustness of frequency-wise feature disentangling. The proposed method obtains state-of-the-art performance on three widely used domain generalization benchmarks (Digit-DG, Office-Home, and PACS).

arXiv:2201.08017공식 코드 · Papers with Code

Fine-Grained Trajectory-based Travel Time Estimation for Multi-city Scenarios Based on Deep Meta-Learning

Travel Time Estimation (TTE) is indispensable in intelligent transportation system (ITS). It is significant to achieve the fine-grained Trajectory-based Travel Time Estimation (TTTE) for multi-city scenarios, namely to accurately estimate travel time of the given trajectory for multiple city scenarios. However, it faces great challenges due to complex factors including dynamic temporal dependencies and fine-grained spatial dependencies. To tackle these challenges, we propose a meta learning based framework, MetaTTE, to continuously provide accurate travel time estimation over time by leveraging well-designed deep neural network model called DED, which consists of Data preprocessing module and Encoder-Decoder network module. By introducing meta learning techniques, the generalization ability of MetaTTE is enhanced using small amount of examples, which opens up new opportunities to increase the potential of achieving consistent performance on TTTE when traffic conditions and road networks change over time in the future. The DED model adopts an encoder-decoder network to capture fine-grained spatial and temporal representations. Extensive experiments on two real-world datasets are conducted to confirm that our MetaTTE outperforms six state-of-art baselines, and improve 29.35% and 25.93% accuracy than the best baseline on Chengdu and Porto datasets, respectively.

arXiv:2201.07753

Deep Capsule Encoder-Decoder Network for Surrogate Modeling and Uncertainty Quantification

We propose a novel \textit{capsule} based deep encoder-decoder model for surrogate modeling and uncertainty quantification of systems in mechanics from sparse data. The proposed framework is developed by adapting Capsule Network (CapsNet) architecture into image-to-image regression encoder-decoder network. Specifically, the aim is to exploit the benefits of CapsNet over convolution neural network (CNN) $-$ retaining pose and position information related to an entity to name a few. The performance of proposed approach is illustrated by solving an elliptic stochastic partial differential equation (SPDE), which also governs systems in mechanics such as steady heat conduction, ground water flow or other diffusion processes, based uncertainty quantification problem with an input dimensionality of $1024$. However, the problem definition does not the restrict the random diffusion field to a particular covariance structure, and the more strenuous task of response prediction for an arbitrary diffusion field is solved. The obtained results from performance evaluation indicate that the proposed approach is accurate, efficient, and robust.