PyTorchKR Papers

최신 논문

총 5,358편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2208.14357

Compound Figure Separation of Biomedical Images: Mining Large Datasets for Self-supervised Learning

With the rapid development of self-supervised learning (e.g., contrastive learning), the importance of having large-scale images (even without annotations) for training a more generalizable AI model has been widely recognized in medical image analysis. However, collecting large-scale task-specific unannotated data at scale can be challenging for individual labs. Existing online resources, such as digital books, publications, and search engines, provide a new resource for obtaining large-scale images. However, published images in healthcare (e.g., radiology and pathology) consist of a considerable amount of compound figures with subplots. In order to extract and separate compound figures into usable individual images for downstream learning, we propose a simple compound figure separation (SimCFS) framework without using the traditionally required detection bounding box annotations, with a new loss function and a hard case simulation. Our technical contribution is four-fold: (1) we introduce a simulation-based training framework that minimizes the need for resource extensive bounding box annotations; (2) we propose a new side loss that is optimized for compound figure separation; (3) we propose an intra-class image augmentation method to simulate hard cases; and (4) to the best of our knowledge, this is the first study that evaluates the efficacy of leveraging self-supervised learning with compound image separation. From the results, the proposed SimCFS achieved state-of-the-art performance on the ImageCLEF 2016 Compound Figure Separation Database. The pretrained self-supervised learning model using large-scale mined figures improved the accuracy of downstream image classification tasks with a contrastive learning algorithm. The source code of SimCFS is made publicly available at https://github.com/hrlblab/ImageSeperation.

arXiv:2208.13721공식 코드 · Papers with Code

CounTR: Transformer-based Generalised Visual Counting

In this paper, we consider the problem of generalised visual object counting, with the goal of developing a computational model for counting the number of objects from arbitrary semantic categories, using arbitrary number of "exemplars", i.e. zero-shot or few-shot counting. To this end, we make the following four contributions: (1) We introduce a novel transformer-based architecture for generalised visual object counting, termed as Counting Transformer (CounTR), which explicitly capture the similarity between image patches or with given "exemplars" with the attention mechanism;(2) We adopt a two-stage training regime, that first pre-trains the model with self-supervised learning, and followed by supervised fine-tuning;(3) We propose a simple, scalable pipeline for synthesizing training images with a large number of instances or that from different semantic categories, explicitly forcing the model to make use of the given "exemplars";(4) We conduct thorough ablation studies on the large-scale counting benchmark, e.g. FSC-147, and demonstrate state-of-the-art performance on both zero and few-shot settings.

arXiv:2208.12666

Effectiveness of Mining Audio and Text Pairs from Public Data for Improving ASR Systems for Low-Resource Languages

End-to-end (E2E) models have become the default choice for state-of-the-art speech recognition systems. Such models are trained on large amounts of labelled data, which are often not available for low-resource languages. Techniques such as self-supervised learning and transfer learning hold promise, but have not yet been effective in training accurate models. On the other hand, collecting labelled datasets on a diverse set of domains and speakers is very expensive. In this work, we demonstrate an inexpensive and effective alternative to these approaches by ``mining'' text and audio pairs for Indian languages from public sources, specifically from the public archives of All India Radio. As a key component, we adapt the Needleman-Wunsch algorithm to align sentences with corresponding audio segments given a long audio and a PDF of its transcript, while being robust to errors due to OCR, extraneous text, and non-transcribed speech. We thus create Shrutilipi, a dataset which contains over 6,400 hours of labelled audio across 12 Indian languages totalling to 4.95M sentences. On average, Shrutilipi results in a 2.3x increase over publicly available labelled data. We establish the quality of Shrutilipi with 21 human evaluators across the 12 languages. We also establish the diversity of Shrutilipi in terms of represented regions, speakers, and mentioned named entities. Significantly, we show that adding Shrutilipi to the training set of Wav2Vec models leads to an average decrease in WER of 5.8\% for 7 languages on the IndicSUPERB benchmark. For Hindi, which has the most benchmarks (7), the average WER falls from 18.8% to 13.5%. This improvement extends to efficient models: We show a 2.3% drop in WER for a Conformer model (10x smaller than Wav2Vec). Finally, we demonstrate the diversity of Shrutilipi by showing that the model trained with it is more robust to noisy input.

arXiv:2208.12263공식 코드 · Papers with Code

Augmenting Reinforcement Learning with Transformer-based Scene Representation Learning for Decision-making of Autonomous Driving

Decision-making for urban autonomous driving is challenging due to the stochastic nature of interactive traffic participants and the complexity of road structures. Although reinforcement learning (RL)-based decision-making scheme is promising to handle urban driving scenarios, it suffers from low sample efficiency and poor adaptability. In this paper, we propose Scene-Rep Transformer to improve the RL decision-making capabilities with better scene representation encoding and sequential predictive latent distillation. Specifically, a multi-stage Transformer (MST) encoder is constructed to model not only the interaction awareness between the ego vehicle and its neighbors but also intention awareness between the agents and their candidate routes. A sequential latent Transformer (SLT) with self-supervised learning objectives is employed to distill the future predictive information into the latent scene representation, in order to reduce the exploration space and speed up training. The final decision-making module based on soft actor-critic (SAC) takes as input the refined latent scene representation from the Scene-Rep Transformer and outputs driving actions. The framework is validated in five challenging simulated urban scenarios with dense traffic, and its performance is manifested quantitatively by the substantial improvements in data efficiency and performance in terms of success rate, safety, and efficiency. The qualitative results reveal that our framework is able to extract the intentions of neighbor agents to help make decisions and deliver more diversified driving behaviors.

arXiv:2208.12345공식 코드 · Papers with Code

Light-weight probing of unsupervised representations for Reinforcement Learning

Unsupervised visual representation learning offers the opportunity to leverage large corpora of unlabeled trajectories to form useful visual representations, which can benefit the training of reinforcement learning (RL) algorithms. However, evaluating the fitness of such representations requires training RL algorithms which is computationally intensive and has high variance outcomes. Inspired by the vision community, we study whether linear probing can be a proxy evaluation task for the quality of unsupervised RL representation. Specifically, we probe for the observed reward in a given state and the action of an expert in a given state, both of which are generally applicable to many RL domains. Through rigorous experimentation, we show that the probing tasks are strongly rank correlated with the downstream RL performance on the Atari100k Benchmark, while having lower variance and up to 600x lower computational cost. This provides a more efficient method for exploring the space of pretraining algorithms and identifying promising pretraining recipes without the need to run RL evaluations for every setting. Leveraging this framework, we further improve existing self-supervised learning (SSL) recipes for RL, highlighting the importance of the forward model, the size of the visual backbone, and the precise formulation of the unsupervised objective.

arXiv:2208.11821공식 코드 · Papers with Code

Refine and Represent: Region-to-Object Representation Learning

Recent works in self-supervised learning have demonstrated strong performance on scene-level dense prediction tasks by pretraining with object-centric or region-based correspondence objectives. In this paper, we present Region-to-Object Representation Learning (R2O) which unifies region-based and object-centric pretraining. R2O operates by training an encoder to dynamically refine region-based segments into object-centric masks and then jointly learns representations of the contents within the mask. R2O uses a "region refinement module" to group small image regions, generated using a region-level prior, into larger regions which tend to correspond to objects by clustering region-level features. As pretraining progresses, R2O follows a region-to-object curriculum which encourages learning region-level features early on and gradually progresses to train object-centric representations. Representations learned using R2O lead to state-of-the art performance in semantic segmentation for PASCAL VOC (+0.7 mIOU) and Cityscapes (+0.4 mIOU) and instance segmentation on MS COCO (+0.3 mask AP). Further, after pretraining on ImageNet, R2O pretrained models are able to surpass existing state-of-the-art in unsupervised object segmentation on the Caltech-UCSD Birds 200-2011 dataset (+2.9 mIoU) without any further training. We provide the code/models from this work at https://github.com/KKallidromitis/r2o.

arXiv:2208.12160

Clustering Egocentric Images in Passive Dietary Monitoring with Self-Supervised Learning

In our recent dietary assessment field studies on passive dietary monitoring in Ghana, we have collected over 250k in-the-wild images. The dataset is an ongoing effort to facilitate accurate measurement of individual food and nutrient intake in low and middle income countries with passive monitoring camera technologies. The current dataset involves 20 households (74 subjects) from both the rural and urban regions of Ghana, and two different types of wearable cameras were used in the studies. Once initiated, wearable cameras continuously capture subjects' activities, which yield massive amounts of data to be cleaned and annotated before analysis is conducted. To ease the data post-processing and annotation tasks, we propose a novel self-supervised learning framework to cluster the large volume of egocentric images into separate events. Each event consists of a sequence of temporally continuous and contextually similar images. By clustering images into separate events, annotators and dietitians can examine and analyze the data more efficiently and facilitate the subsequent dietary assessment processes. Validated on a held-out test set with ground truth labels, the proposed framework outperforms baselines in terms of clustering quality and classification accuracy.

arXiv:2208.11278

Federated Self-Supervised Contrastive Learning and Masked Autoencoder for Dermatological Disease Diagnosis

In dermatological disease diagnosis, the private data collected by mobile dermatology assistants exist on distributed mobile devices of patients. Federated learning (FL) can use decentralized data to train models while keeping data local. Existing FL methods assume all the data have labels. However, medical data often comes without full labels due to high labeling costs. Self-supervised learning (SSL) methods, contrastive learning (CL) and masked autoencoders (MAE), can leverage the unlabeled data to pre-train models, followed by fine-tuning with limited labels. However, combining SSL and FL has unique challenges. For example, CL requires diverse data but each device only has limited data. For MAE, while Vision Transformer (ViT) based MAE has higher accuracy over CNNs in centralized learning, MAE's performance in FL with unlabeled data has not been investigated. Besides, the ViT synchronization between the server and clients is different from traditional CNNs. Therefore, special synchronization methods need to be designed. In this work, we propose two federated self-supervised learning frameworks for dermatological disease diagnosis with limited labels. The first one features lower computation costs, suitable for mobile devices. The second one features high accuracy and fits high-performance servers. Based on CL, we proposed federated contrastive learning with feature sharing (FedCLF). Features are shared for diverse contrastive information without sharing raw data for privacy. Based on MAE, we proposed FedMAE. Knowledge split separates the global and local knowledge learned from each client. Only global knowledge is aggregated for higher generalization performance. Experiments on dermatological disease datasets show superior accuracy of the proposed frameworks over state-of-the-arts.

arXiv:2208.11435

Bidirectional Contrastive Split Learning for Visual Question Answering

Visual Question Answering (VQA) based on multi-modal data facilitates real-life applications such as home robots and medical diagnoses. One significant challenge is to devise a robust decentralized learning framework for various client models where centralized data collection is refrained due to confidentiality concerns. This work aims to tackle privacy-preserving VQA by decoupling a multi-modal model into representation modules and a contrastive module and leveraging inter-module gradients sharing and inter-client weight sharing. To this end, we propose Bidirectional Contrastive Split Learning (BiCSL) to train a global multi-modal model on the entire data distribution of decentralized clients. We employ the contrastive loss that enables a more efficient self-supervised learning of decentralized modules. Comprehensive experiments are conducted on the VQA-v2 dataset based on five SOTA VQA models, demonstrating the effectiveness of the proposed method. Furthermore, we inspect BiCSL's robustness against a dual-key backdoor attack on VQA. Consequently, BiCSL shows much better robustness to the multi-modal adversarial attack compared to the centralized learning method, which provides a promising approach to decentralized multi-modal learning.

arXiv:2208.11424공식 코드 · Papers with Code

Self-Supervised Endoscopic Image Key-Points Matching

Feature matching and finding correspondences between endoscopic images is a key step in many clinical applications such as patient follow-up and generation of panoramic image from clinical sequences for fast anomalies localization. Nonetheless, due to the high texture variability present in endoscopic images, the development of robust and accurate feature matching becomes a challenging task. Recently, deep learning techniques which deliver learned features extracted via convolutional neural networks (CNNs) have gained traction in a wide range of computer vision tasks. However, they all follow a supervised learning scheme where a large amount of annotated data is required to reach good performances, which is generally not always available for medical data databases. To overcome this limitation related to labeled data scarcity, the self-supervised learning paradigm has recently shown great success in a number of applications. This paper proposes a novel self-supervised approach for endoscopic image matching based on deep learning techniques. When compared to standard hand-crafted local feature descriptors, our method outperformed them in terms of precision and recall. Furthermore, our self-supervised descriptor provides a competitive performance in comparison to a selection of state-of-the-art deep learning based supervised methods in terms of precision and matching score.

arXiv:2208.10893

Transfer Learning Application of Self-supervised Learning in ARPES

Recent development in angle-resolved photoemission spectroscopy (ARPES) technique involves spatially resolving samples while maintaining the high-resolution feature of momentum space. This development easily expands the data size and its complexity for data analysis, where one of it is to label similar dispersion cuts and map them spatially. In this work, we demonstrate that the recent development in representational learning (self-supervised learning) model combined with k-means clustering can help automate that part of data analysis and save precious time, albeit with low performance. Finally, we introduce a few-shot learning (k-nearest neighbour or kNN) in representational space where we selectively choose one (k=1) image reference for each known label and subsequently label the rest of the data with respect to the nearest reference image. This last approach demonstrates the strength of the self-supervised learning to automate the image analysis in ARPES in particular and can be generalized into any science data analysis that heavily involves image data.

arXiv:2208.11052공식 코드 · Papers with Code

IMPaSh: A Novel Domain-shift Resistant Representation for Colorectal Cancer Tissue Classification

The appearance of histopathology images depends on tissue type, staining and digitization procedure. These vary from source to source and are the potential causes for domain-shift problems. Owing to this problem, despite the great success of deep learning models in computational pathology, a model trained on a specific domain may still perform sub-optimally when we apply them to another domain. To overcome this, we propose a new augmentation called PatchShuffling and a novel self-supervised contrastive learning framework named IMPaSh for pre-training deep learning models. Using these, we obtained a ResNet50 encoder that can extract image representation resistant to domain-shift. We compared our derived representation against those acquired based on other domain-generalization techniques by using them for the cross-domain classification of colorectal tissue images. We show that the proposed method outperforms other traditional histology domain-adaptation and state-of-the-art self-supervised learning methods. Code is available at: https://github.com/trinhvg/IMPash .

arXiv:2208.10779공식 코드 · Papers with Code

Efficient Self-Supervision using Patch-based Contrastive Learning for Histopathology Image Segmentation

Learning discriminative representations of unlabelled data is a challenging task. Contrastive self-supervised learning provides a framework to learn meaningful representations using learned notions of similarity measures from simple pretext tasks. In this work, we propose a simple and efficient framework for self-supervised image segmentation using contrastive learning on image patches, without using explicit pretext tasks or any further labeled fine-tuning. A fully convolutional neural network (FCNN) is trained in a self-supervised manner to discern features in the input images and obtain confidence maps which capture the network's belief about the objects belonging to the same class. Positive- and negative- patches are sampled based on the average entropy in the confidence maps for contrastive learning. Convergence is assumed when the information separation between the positive patches is small, and the positive-negative pairs is large. The proposed model only consists of a simple FCNN with 10.8k parameters and requires about 5 minutes to converge on the high resolution microscopy datasets, which is orders of magnitude smaller than the relevant self-supervised methods to attain similar performance. We evaluate the proposed method for the task of segmenting nuclei from two histopathology datasets, and show comparable performance with relevant self-supervised and supervised methods.

arXiv:2208.10769

PIFu for the Real World: A Self-supervised Framework to Reconstruct Dressed Human from Single-view Images

It is very challenging to accurately reconstruct sophisticated human geometry caused by various poses and garments from a single image. Recently, works based on pixel-aligned implicit function (PIFu) have made a big step and achieved state-of-the-art fidelity on image-based 3D human digitization. However, the training of PIFu relies heavily on expensive and limited 3D ground truth data (i.e. synthetic data), thus hindering its generalization to more diverse real world images. In this work, we propose an end-to-end self-supervised network named SelfPIFu to utilize abundant and diverse in-the-wild images, resulting in largely improved reconstructions when tested on unconstrained in-the-wild images. At the core of SelfPIFu is the depth-guided volume-/surface-aware signed distance fields (SDF) learning, which enables self-supervised learning of a PIFu without access to GT mesh. The whole framework consists of a normal estimator, a depth estimator, and a SDF-based PIFu and better utilizes extra depth GT during training. Extensive experiments demonstrate the effectiveness of our self-supervised framework and the superiority of using depth as input. On synthetic data, our Intersection-Over-Union (IoU) achieves to 93.5%, 18% higher compared with PIFuHD. For in-the-wild images, we conduct user studies on the reconstructed results, the selection rate of our results is over 68% compared with other state-of-the-art methods.

arXiv:2208.10694

Spiral Contrastive Learning: An Efficient 3D Representation Learning Method for Unannotated CT Lesions

Computed tomography (CT) samples with pathological annotations are difficult to obtain. As a result, the computer-aided diagnosis (CAD) algorithms are trained on small datasets (e.g., LIDC-IDRI with 1,018 samples), limiting their accuracies and reliability. In the past five years, several works have tailored for unsupervised representations of CT lesions via two-dimensional (2D) and three-dimensional (3D) self-supervised learning (SSL) algorithms. The 2D algorithms have difficulty capturing 3D information, and existing 3D algorithms are computationally heavy. Light-weight 3D SSL remains the boundary to explore. In this paper, we propose the spiral contrastive learning (SCL), which yields 3D representations in a computationally efficient manner. SCL first transforms 3D lesions to the 2D plane using an information-preserving spiral transformation, and then learn transformation-invariant features using 2D contrastive learning. For the augmentation, we consider natural image augmentations and medical image augmentations. We evaluate SCL by training a classification head upon the embedding layer. Experimental results show that SCL achieves state-of-the-art accuracy on LIDC-IDRI (89.72%), LNDb (82.09%) and TianChi (90.16%) for unsupervised representation learning. With 10% annotated data for fine-tune, the performance of SCL is comparable to that of supervised learning algorithms (85.75% vs. 85.03% on LIDC-IDRI, 78.20% vs. 73.44% on LNDb and 87.85% vs. 83.34% on TianChi, respectively). Meanwhile, SCL reduces the computational effort by 66.98% compared to other 3D SSL algorithms, demonstrating the efficiency of the proposed method in unsupervised pre-training.

arXiv:2208.10493공식 코드 · Papers with Code

Relational Self-Supervised Learning on Graphs

Over the past few years, graph representation learning (GRL) has been a powerful strategy for analyzing graph-structured data. Recently, GRL methods have shown promising results by adopting self-supervised learning methods developed for learning representations of images. Despite their success, existing GRL methods tend to overlook an inherent distinction between images and graphs, i.e., images are assumed to be independently and identically distributed, whereas graphs exhibit relational information among data instances, i.e., nodes. To fully benefit from the relational information inherent in the graph-structured data, we propose a novel GRL method, called RGRL, that learns from the relational information generated from the graph itself. RGRL learns node representations such that the relationship among nodes is invariant to augmentations, i.e., augmentation-invariant relationship, which allows the node representations to vary as long as the relationship among the nodes is preserved. By considering the relationship among nodes in both global and local perspectives, RGRL overcomes limitations of previous contrastive and non-contrastive methods, and achieves the best of both worlds. Extensive experiments on fourteen benchmark datasets over various downstream tasks demonstrate the superiority of RGRL over state-of-the-art baselines. The source code for RGRL is available at https://github.com/Namkyeong/RGRL.

arXiv:2208.10061공식 코드 · Papers with Code

Improving Knowledge-aware Recommendation with Multi-level Interactive Contrastive Learning

Incorporating Knowledge Graphs (KG) into recommeder system has attracted considerable attention. Recently, the technical trend of Knowledge-aware Recommendation (KGR) is to develop end-to-end models based on graph neural networks (GNNs). However, the extremely sparse user-item interactions significantly degrade the performance of the GNN-based models, as: 1) the sparse interaction, means inadequate supervision signals and limits the supervised GNN-based models; 2) the combination of sparse interactions (CF part) and redundant KG facts (KG part) results in an unbalanced information utilization. Besides, the GNN paradigm aggregates local neighbors for node representation learning, while ignoring the non-local KG facts and making the knowledge extraction insufficient. Inspired by the recent success of contrastive learning in mining supervised signals from data itself, in this paper, we focus on exploring contrastive learning in KGR and propose a novel multi-level interactive contrastive learning mechanism. Different from traditional contrastive learning methods which contrast nodes of two generated graph views, interactive contrastive mechanism conducts layer-wise self-supervised learning by contrasting layers of different parts within graphs, which is also an "interaction" action. Specifically, we first construct local and non-local graphs for user/item in KG, exploring more KG facts for KGR. Then an intra-graph level interactive contrastive learning is performed within each graph, which contrasts layers of the CF and KG parts, for more consistent information leveraging. Besides, an inter-graph level interactive contrastive learning is performed between the local and non-local graphs, for sufficiently and coherently extracting non-local KG signals. Extensive experiments conducted on three benchmark datasets show the superior performance of our proposed method over the state-of-the-arts.

arXiv:2208.09957공식 코드 · Papers with Code

Heterogeneous Graph Masked Autoencoders

Generative self-supervised learning (SSL), especially masked autoencoders, has become one of the most exciting learning paradigms and has shown great potential in handling graph data. However, real-world graphs are always heterogeneous, which poses three critical challenges that existing methods ignore: 1) how to capture complex graph structure? 2) how to incorporate various node attributes? and 3) how to encode different node positions? In light of this, we study the problem of generative SSL on heterogeneous graphs and propose HGMAE, a novel heterogeneous graph masked autoencoder model to address these challenges. HGMAE captures comprehensive graph information via two innovative masking techniques and three unique training strategies. In particular, we first develop metapath masking and adaptive attribute masking with dynamic mask rate to enable effective and stable learning on heterogeneous graphs. We then design several training strategies including metapath-based edge reconstruction to adopt complex structural information, target attribute restoration to incorporate various node attributes, and positional feature prediction to encode node positional information. Extensive experiments demonstrate that HGMAE outperforms both contrastive and generative state-of-the-art baselines on several tasks across multiple datasets. Codes are available at https://github.com/meettyj/HGMAE.

arXiv:2208.09849

Semantic-Enhanced Image Clustering

Image clustering is an important and open-challenging task in computer vision. Although many methods have been proposed to solve the image clustering task, they only explore images and uncover clusters according to the image features, thus being unable to distinguish visually similar but semantically different images. In this paper, we propose to investigate the task of image clustering with the help of a visual-language pre-training model. Different from the zero-shot setting, in which the class names are known, we only know the number of clusters in this setting. Therefore, how to map images to a proper semantic space and how to cluster images from both image and semantic spaces are two key problems. To solve the above problems, we propose a novel image clustering method guided by the visual-language pre-training model CLIP, named \textbf{Semantic-Enhanced Image Clustering (SIC)}. In this new method, we propose a method to map the given images to a proper semantic space first and efficient methods to generate pseudo-labels according to the relationships between images and semantics. Finally, we propose performing clustering with consistency learning in both image space and semantic space, in a self-supervised learning fashion. The theoretical result of convergence analysis shows that our proposed method can converge at a sublinear speed. Theoretical analysis of expectation risk also shows that we can reduce the expected risk by improving neighborhood consistency, increasing prediction confidence, or reducing neighborhood imbalance. Experimental results on five benchmark datasets clearly show the superiority of our new method.

arXiv:2208.09668

Generalised Co-Salient Object Detection

We propose a new setting that relaxes an assumption in the conventional Co-Salient Object Detection (CoSOD) setting by allowing the presence of "noisy images" which do not show the shared co-salient object. We call this new setting Generalised Co-Salient Object Detection (GCoSOD). We propose a novel random sampling based Generalised CoSOD Training (GCT) strategy to distill the awareness of inter-image absence of co-salient objects into CoSOD models. It employs a Diverse Sampling Self-Supervised Learning (DS3L) that, in addition to the provided supervised co-salient label, introduces additional self-supervised labels for noisy images (being null, that no co-salient object is present). Further, the random sampling process inherent in GCT enables the generation of a high-quality uncertainty map highlighting potential false-positive predictions at instance level. To evaluate the performance of CoSOD models under the GCoSOD setting, we propose two new testing datasets, namely CoCA-Common and CoCA-Zero, where a common salient object is partially present in the former and completely absent in the latter. Extensive experiments demonstrate that our proposed method significantly improves the performance of CoSOD models in terms of the performance under the GCoSOD setting as well as the model calibration degrees.