PyTorchKR Papers

최신 논문

총 716편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2104.09864공식 코드 · Papers with Code

RoFormer: Enhanced Transformer with Rotary Position Embedding

Position encoding recently has shown effective in the transformer architecture. It enables valuable supervision for dependency modeling between elements at different positions of the sequence. In this paper, we first investigate various methods to integrate positional information into the learning process of transformer-based language models. Then, we propose a novel method named Rotary Position Embedding(RoPE) to effectively leverage the positional information. Specifically, the proposed RoPE encodes the absolute position with a rotation matrix and meanwhile incorporates the explicit relative position dependency in self-attention formulation. Notably, RoPE enables valuable properties, including the flexibility of sequence length, decaying inter-token dependency with increasing relative distances, and the capability of equipping the linear self-attention with relative position encoding. Finally, we evaluate the enhanced transformer with rotary position embedding, also called RoFormer, on various long text classification benchmark datasets. Our experiments show that it consistently overcomes its alternatives. Furthermore, we provide a theoretical analysis to explain some experimental results. RoFormer is already integrated into Huggingface: \url{https://huggingface.co/docs/transformers/model_doc/roformer}.

arXiv:2012.06142공식 코드 · Papers with Code

Iterative Geometry Calibration from Distance Estimates for Wireless Acoustic Sensor Networks

In this paper we present an approach to geometry calibration in wireless acoustic sensor networks, whose nodes are assumed to be equipped with a compact microphone array. The proposed approach solely works with estimates of the distances between acoustic sources and the nodes that record these sources. It consists of an iterative weighted least squares localization procedure, which is initialized by multidimensional scaling. Alongside the sensor node locations, also the positions of the acoustic sources are estimated. Furthermore, we derive the Cramer-Rao lower bound (CRLB) for source and sensor position estimation, and show by simulation that the estimator is efficient.

arXiv:2104.06728공식 코드 · Papers with Code

Adversarial Sticker: A Stealthy Attack Method in the Physical World

To assess the vulnerability of deep learning in the physical world, recent works introduce adversarial patches and apply them on different tasks. In this paper, we propose another kind of adversarial patch: the Meaningful Adversarial Sticker, a physically feasible and stealthy attack method by using real stickers existing in our life. Unlike the previous adversarial patches by designing perturbations, our method manipulates the sticker's pasting position and rotation angle on the objects to perform physical attacks. Because the position and rotation angle are less affected by the printing loss and color distortion, adversarial stickers can keep good attacking performance in the physical world. Besides, to make adversarial stickers more practical in real scenes, we conduct attacks in the black-box setting with the limited information rather than the white-box setting with all the details of threat models. To effectively solve for the sticker's parameters, we design the Region based Heuristic Differential Evolution Algorithm, which utilizes the new-found regional aggregation of effective solutions and the adaptive adjustment strategy of the evaluation criteria. Our method is comprehensively verified in the face recognition and then extended to the image retrieval and traffic sign recognition. Extensive experiments show the proposed method is effective and efficient in complex physical conditions and has a good generalization for different tasks.

arXiv:2104.00380공식 코드 · Papers with Code

Online Multiple Object Tracking with Cross-Task Synergy

Modern online multiple object tracking (MOT) methods usually focus on two directions to improve tracking performance. One is to predict new positions in an incoming frame based on tracking information from previous frames, and the other is to enhance data association by generating more discriminative identity embeddings. Some works combined both directions within one framework but handled them as two individual tasks, thus gaining little mutual benefits. In this paper, we propose a novel unified model with synergy between position prediction and embedding association. The two tasks are linked by temporal-aware target attention and distractor attention, as well as identity-aware memory aggregation model. Specifically, the attention modules can make the prediction focus more on targets and less on distractors, therefore more reliable embeddings can be extracted accordingly for association. On the other hand, such reliable embeddings can boost identity-awareness through memory aggregation, hence strengthen attention modules and suppress drifts. In this way, the synergy between position prediction and embedding association is achieved, which leads to strong robustness to occlusions. Extensive experiments demonstrate the superiority of our proposed model over a wide range of existing methods on MOTChallenge benchmarks. Our code and models are publicly available at https://github.com/songguocode/TADAM.

arXiv:2005.07108공식 코드 · Papers with Code

Effective Model of Loop Extrusion Predicts Chromosomal Domains

An active loop-extrusion mechanism is regarded as the main out--of--equilibrium mechanism responsible for the structuring of megabase-sized domains in chromosomes. We developed a model to study the dynamics of the chromosome fibre by solving the kinetic equations associated with the motion of the extruder. By averaging out the position of the extruder along the chain, we build an effective equilibrium model capable of reproducing experimental contact maps based solely on the positions of extrusion--blocking proteins. We assessed the quality of the effective model using numerical simulations of chromosomal segments and comparing the results with explicit-extruder models and experimental data.

arXiv:2103.15255공식 코드 · Papers with Code

A More Fine-Grained Aspect-Sentiment-Opinion Triplet Extraction Task

Aspect Sentiment Triplet Extraction (ASTE) aims to extract aspect term, sentiment and opinion term triplets from sentences and tries to provide a complete solution for aspect-based sentiment analysis (ABSA). However, some triplets extracted by ASTE are confusing, since the sentiment in a triplet extracted by ASTE is the sentiment that the sentence expresses toward the aspect term rather than the sentiment of the aspect term and opinion term pair. In this paper, we introduce a more fine-grained Aspect-Sentiment-Opinion Triplet Extraction (ASOTE) Task. ASOTE also extracts aspect term, sentiment and opinion term triplets. However, the sentiment in a triplet extracted by ASOTE is the sentiment of the aspect term and opinion term pair. We build four datasets for ASOTE based on several popular ABSA benchmarks. We propose a Position-aware BERT-based Framework (PBF) to address this task. PBF first extracts aspect terms from sentences. For each extracted aspect term, PBF first generates aspect term-specific sentence representations considering both the meaning and the position of the aspect term, then extracts associated opinion terms and predicts the sentiments of the aspect term and opinion term pairs based on the sentence representations. Experimental results on the four datasets show the effectiveness of PBF.

arXiv:2103.14635공식 코드 · Papers with Code

PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds

We introduce Position Adaptive Convolution (PAConv), a generic convolution operation for 3D point cloud processing. The key of PAConv is to construct the convolution kernel by dynamically assembling basic weight matrices stored in Weight Bank, where the coefficients of these weight matrices are self-adaptively learned from point positions through ScoreNet. In this way, the kernel is built in a data-driven manner, endowing PAConv with more flexibility than 2D convolutions to better handle the irregular and unordered point cloud data. Besides, the complexity of the learning process is reduced by combining weight matrices instead of brutally predicting kernels from point positions. Furthermore, different from the existing point convolution operators whose network architectures are often heavily engineered, we integrate our PAConv into classical MLP-based point cloud pipelines without changing network configurations. Even built on simple networks, our method still approaches or even surpasses the state-of-the-art models, and significantly improves baseline performance on both classification and segmentation tasks, yet with decent efficiency. Thorough ablation studies and visualizations are provided to understand PAConv. Code is released on https://github.com/CVMI-Lab/PAConv.

arXiv:2103.09030공식 코드 · Papers with Code

A Large-Scale Dataset for Benchmarking Elevator Button Segmentation and Character Recognition

Human activities are hugely restricted by COVID-19, recently. Robots that can conduct inter-floor navigation attract much public attention, since they can substitute human workers to conduct the service work. However, current robots either depend on human assistance or elevator retrofitting, and fully autonomous inter-floor navigation is still not available. As the very first step of inter-floor navigation, elevator button segmentation and recognition hold an important position. Therefore, we release the first large-scale publicly available elevator panel dataset in this work, containing 3,718 panel images with 35,100 button labels, to facilitate more powerful algorithms on autonomous elevator operation. Together with the dataset, a number of deep learning based implementations for button segmentation and recognition are also released to benchmark future methods in the community. The dataset will be available at \url{https://github.com/zhudelong/elevator_button_recognition

arXiv:2103.06879공식 코드 · Papers with Code

CoMoGAN: continuous model-guided image-to-image translation

CoMoGAN is a continuous GAN relying on the unsupervised reorganization of the target data on a functional manifold. To that matter, we introduce a new Functional Instance Normalization layer and residual mechanism, which together disentangle image content from position on target manifold. We rely on naive physics-inspired models to guide the training while allowing private model/translations features. CoMoGAN can be used with any GAN backbone and allows new types of image translation, such as cyclic image translation like timelapse generation, or detached linear translation. On all datasets, it outperforms the literature. Our code is available at http://github.com/cv-rits/CoMoGAN .

arXiv:2103.06725공식 코드 · Papers with Code

Duplex Contextual Relation Network for Polyp Segmentation

Polyp segmentation is of great importance in the early diagnosis and treatment of colorectal cancer. Since polyps vary in their shape, size, color, and texture, accurate polyp segmentation is very challenging. One promising way to mitigate the diversity of polyps is to model the contextual relation for each pixel such as using attention mechanism. However, previous methods only focus on learning the dependencies between the position within an individual image and ignore the contextual relation across different images. In this paper, we propose Duplex Contextual Relation Network (DCRNet) to capture both within-image and cross-image contextual relations. Specifically, we first design Interior Contextual-Relation Module to estimate the similarity between each position and all the positions within the same image. Then Exterior Contextual-Relation Module is incorporated to estimate the similarity between each position and the positions across different images. Based on the above two types of similarity, the feature at one position can be further enhanced by the contextual region embedding within and across images. To store the characteristic region embedding from all the images, a memory bank is designed and operates as a queue. Therefore, the proposed method can relate similar features even though they come from different images. We evaluate the proposed method on the EndoScene, Kvasir-SEG and the recently released large-scale PICCOLO dataset. Experimental results show that the proposed DCRNet outperforms the state-of-the-art methods in terms of the widely-used evaluation metrics.

arXiv:2103.04180공식 코드 · Papers with Code

Neural networks can understand compositional functions that humans do not, in the context of emergent communication

We show that it is possible to craft transformations that, applied to compositional grammars, result in grammars that neural networks can learn easily, but humans do not. This could explain the disconnect between current metrics of compositionality, that are arguably human-centric, and the ability of neural networks to generalize to unseen examples. We propose to use the transformations as a benchmark, ICY, which could be used to measure aspects of the compositional inductive bias of networks, and to search for networks with similar compositional inductive biases to humans. As an example of this approach, we propose a hierarchical model, HU-RNN, which shows an inductive bias towards position-independent, word-like groups of tokens.

arXiv:2103.02845공식 코드 · Papers with Code

Camera-Space Hand Mesh Recovery via Semantic Aggregation and Adaptive 2D-1D Registration

Recent years have witnessed significant progress in 3D hand mesh recovery. Nevertheless, because of the intrinsic 2D-to-3D ambiguity, recovering camera-space 3D information from a single RGB image remains challenging. To tackle this problem, we divide camera-space mesh recovery into two sub-tasks, i.e., root-relative mesh recovery and root recovery. First, joint landmarks and silhouette are extracted from a single input image to provide 2D cues for the 3D tasks. In the root-relative mesh recovery task, we exploit semantic relations among joints to generate a 3D mesh from the extracted 2D cues. Such generated 3D mesh coordinates are expressed relative to a root position, i.e., wrist of the hand. In the root recovery task, the root position is registered to the camera space by aligning the generated 3D mesh back to 2D cues, thereby completing cameraspace 3D mesh recovery. Our pipeline is novel in that (1) it explicitly makes use of known semantic relations among joints and (2) it exploits 1D projections of the silhouette and mesh to achieve robust registration. Extensive experiments on popular datasets such as FreiHAND, RHD, and Human3.6M demonstrate that our approach achieves stateof-the-art performance on both root-relative mesh recovery and root recovery. Our code is publicly available at https://github.com/SeanChenxy/HandMesh.

arXiv:2102.11588공식 코드 · Papers with Code

Data Fusion for Audiovisual Speaker Localization: Extending Dynamic Stream Weights to the Spatial Domain

Estimating the positions of multiple speakers can be helpful for tasks like automatic speech recognition or speaker diarization. Both applications benefit from a known speaker position when, for instance, applying beamforming or assigning unique speaker identities. Recently, several approaches utilizing acoustic signals augmented with visual data have been proposed for this task. However, both the acoustic and the visual modality may be corrupted in specific spatial regions, for instance due to poor lighting conditions or to the presence of background noise. This paper proposes a novel audiovisual data fusion framework for speaker localization by assigning individual dynamic stream weights to specific regions in the localization space. This fusion is achieved via a neural network, which combines the predictions of individual audio and video trackers based on their time- and location-dependent reliability. A performance evaluation using audiovisual recordings yields promising results, with the proposed fusion approach outperforming all baseline models.

arXiv:2102.10882공식 코드 · Papers with Code

Conditional Positional Encodings for Vision Transformers

We propose a conditional positional encoding (CPE) scheme for vision Transformers. Unlike previous fixed or learnable positional encodings, which are pre-defined and independent of input tokens, CPE is dynamically generated and conditioned on the local neighborhood of the input tokens. As a result, CPE can easily generalize to the input sequences that are longer than what the model has ever seen during training. Besides, CPE can keep the desired translation-invariance in the image classification task, resulting in improved performance. We implement CPE with a simple Position Encoding Generator (PEG) to get seamlessly incorporated into the current Transformer framework. Built on PEG, we present Conditional Position encoding Vision Transformer (CPVT). We demonstrate that CPVT has visually similar attention maps compared to those with learned positional encodings and delivers outperforming results. Our code is available at https://github.com/Meituan-AutoML/CPVT .

arXiv:2102.10084공식 코드 · Papers with Code

Hate-Alert@DravidianLangTech-EACL2021: Ensembling strategies for Transformer-based Offensive language Detection

Social media often acts as breeding grounds for different forms of offensive content. For low resource languages like Tamil, the situation is more complex due to the poor performance of multilingual or language-specific models and lack of proper benchmark datasets. Based on this shared task, Offensive Language Identification in Dravidian Languages at EACL 2021, we present an exhaustive exploration of different transformer models, We also provide a genetic algorithm technique for ensembling different models. Our ensembled models trained separately for each language secured the first position in Tamil, the second position in Kannada, and the first position in Malayalam sub-tasks. The models and codes are provided.

arXiv:2102.08312공식 코드 · Papers with Code

On Mathews Correlation Coefficient and Improved Distance Map Loss for Automatic Glacier Calving Front Segmentation in SAR Imagery

The vast majority of the outlet glaciers and ice streams of the polar ice sheets end in the ocean. Ice mass loss via calving of the glaciers into the ocean has increased over the last few decades. Information on the temporal variability of the calving front position provides fundamental information on the state of the glacier and ice stream, which can be exploited as calibration and validation data to enhance ice dynamics modeling. To identify the calving front position automatically, deep neural network-based semantic segmentation pipelines can be used to delineate the acquired SAR imagery. However, the extreme class imbalance is highly challenging for the accurate calving front segmentation in these images. Therefore, we propose the use of the Mathews correlation coefficient (MCC) as an early stopping criterion because of its symmetrical properties and its invariance towards class imbalance. Moreover, we propose an improvement to the distance map-based binary cross-entropy (BCE) loss function. The distance map adds context to the loss function about the important regions for segmentation and helps accounting for the imbalanced data. Using Mathews correlation coefficient as early stopping demonstrates an average 15% dice coefficient improvement compared to the commonly used BCE. The modified distance map loss further improves the segmentation performance by another 2%. These results are encouraging as they support the effectiveness of the proposed methods for segmentation problems suffering from extreme class imbalances.

arXiv:2102.07090공식 코드 · Papers with Code

New methods for metastimuli: architecture, embeddings, and neural network optimization

Six significant new methodological developments of the previously-presented "metastimuli architecture" for human learning through machine learning of spatially correlated structural position within a user's personal information management system (PIMS), providing the basis for haptic metastimuli, are presented. These include architectural innovation, recurrent (RNN) artificial neural network (ANN) application, a variety of atom embedding techniques (including a novel technique we call "nabla" embedding inspired by linguistics), ANN hyper-parameter (one that affects the network but is not trained, e.g. the learning rate) optimization, and meta-parameter (one that determines the system performance but is not trained and not a hyper-parameter, e.g. the atom embedding technique) optimization for exploring the large design space. A technique for using the system for automatic atom categorization in a user's PIMS is outlined. ANN training and hyper- and meta-parameter optimization results are presented and discussed in service of methodological recommendations.

arXiv:2102.04024공식 코드 · Papers with Code

IDOL: Inertial Deep Orientation-Estimation and Localization

Many smartphone applications use inertial measurement units (IMUs) to sense movement, but the use of these sensors for pedestrian localization can be challenging due to their noise characteristics. Recent data-driven inertial odometry approaches have demonstrated the increasing feasibility of inertial navigation. However, they still rely upon conventional smartphone orientation estimates that they assume to be accurate, while in fact these orientation estimates can be a significant source of error. To address the problem of inaccurate orientation estimates, we present a two-stage, data-driven pipeline using a commodity smartphone that first estimates device orientations and then estimates device position. The orientation module relies on a recurrent neural network and Extended Kalman Filter to obtain orientation estimates that are used to then rotate raw IMU measurements into the appropriate reference frame. The position module then passes those measurements through another recurrent network architecture to perform localization. Our proposed method outperforms state-of-the-art methods in both orientation and position error on a large dataset we constructed that contains 20 hours of pedestrian motion across 3 buildings and 15 subjects.

arXiv:2102.02898공식 코드 · Papers with Code

Cobiveco: Consistent biventricular coordinates for precise and intuitive description of position in the heart -- with MATLAB implementation

Ventricular coordinates are widely used as a versatile tool for various applications that benefit from a description of local position within the heart. However, the practical usefulness of ventricular coordinates is determined by their ability to meet application-specific requirements. For regression-based estimation of biventricular position, for example, a symmetric definition of coordinate directions in both ventricles is important. For the transfer of data between different hearts as another use case, the consistency of coordinate values across different geometries is particularly relevant. To meet these requirements, we compare different approaches to compute coordinates and present Cobiveco, a symmetric, consistent and intuitive biventricular coordinate system that builds upon existing coordinate systems, but overcomes some of their limitations. A novel one-way transfer error is introduced to assess the consistency of the coordinates. Normalized distances along bijective trajectories between two boundaries were found to be superior to solutions of Laplace's equation for defining coordinate values, as they show better linearity in space. Evaluation of transfer and linearity errors on 36 patient geometries revealed a more than 4-fold improvement compared to a state-of-the-art method. Finally, we show two application examples underlining the relevance for cardiac data processing. Cobiveco MATLAB code is available under a permissive open-source license.

arXiv:2102.02855공식 코드 · Papers with Code

Machine Learning for Auxiliary Sources

We rewrite the numerical ansatz of the Method of Auxiliary Sources (MAS), typically used in computational electromagnetics, as a neural network, i.e. as a composed function of linear and activation layers. MAS is a numerical method for Partial Differential Equations (PDEs) that employs point sources, which are also exact solutions of the considered PDE, as radial basis functions to match a given boundary condition. In the framework of neural networks we rely on optimization algorithms such as Adam to train MAS and find both its optimal coefficients and positions of the central singularities of the sources. In this work we also show that the MAS ansatz trained as a neural network can be used, in the case of an unknown function with a central singularity, to detect the position of such singularity.