PyTorchKR Papers

최신 논문

총 6,724편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2109.15317

Unsupervised Few-Shot Action Recognition via Action-Appearance Aligned Meta-Adaptation

We present MetaUVFS as the first Unsupervised Meta-learning algorithm for Video Few-Shot action recognition. MetaUVFS leverages over 550K unlabeled videos to train a two-stream 2D and 3D CNN architecture via contrastive learning to capture the appearance-specific spatial and action-specific spatio-temporal video features respectively. MetaUVFS comprises a novel Action-Appearance Aligned Meta-adaptation (A3M) module that learns to focus on the action-oriented video features in relation to the appearance features via explicit few-shot episodic meta-learning over unsupervised hard-mined episodes. Our action-appearance alignment and explicit few-shot learner conditions the unsupervised training to mimic the downstream few-shot task, enabling MetaUVFS to significantly outperform all unsupervised methods on few-shot benchmarks. Moreover, unlike previous few-shot action recognition methods that are supervised, MetaUVFS needs neither base-class labels nor a supervised pretrained backbone. Thus, we need to train MetaUVFS just once to perform competitively or sometimes even outperform state-of-the-art supervised methods on popular HMDB51, UCF101, and Kinetics100 few-shot datasets.

arXiv:2109.14157

Hybrid Dynamic Contrast and Probability Distillation for Unsupervised Person Re-Id

Unsupervised person re-identification (Re-Id) has attracted increasing attention due to its practical application in the read-world video surveillance system. The traditional unsupervised Re-Id are mostly based on the method alternating between clustering and fine-tuning with the classification or metric learning objectives on the grouped clusters. However, since person Re-Id is an open-set problem, the clustering based methods often leave out lots of outlier instances or group the instances into the wrong clusters, thus they can not make full use of the training samples as a whole. To solve these problems, we present the hybrid dynamic cluster contrast and probability distillation algorithm. It formulates the unsupervised Re-Id problem into an unified local-to-global dynamic contrastive learning and self-supervised probability distillation framework. Specifically, the proposed method can make the utmost of the self-supervised signals of all the clustered and un-clustered instances, from both the instances' self-contrastive level and the probability distillation respective, in the memory-based non-parametric manner. Besides, the proposed hybrid local-to-global contrastive learning can take full advantage of the informative and valuable training examples for effective and robust training. Extensive experiment results show that the proposed method achieves superior performances to state-of-the-art methods, under both the purely unsupervised and unsupervised domain adaptation experiment settings.

arXiv:2109.14131

Contrastive Video-Language Segmentation

We focus on the problem of segmenting a certain object referred by a natural language sentence in video content, at the core of formulating a pinpoint vision-language relation. While existing attempts mainly construct such relation in an implicit way, i.e., grid-level multi-modal feature fusion, it has been proven problematic to distinguish semantically similar objects under this paradigm. In this work, we propose to interwind the visual and linguistic modalities in an explicit way via the contrastive learning objective, which directly aligns the referred object and the language description and separates the unreferred content apart across frames. Moreover, to remedy for the degradation problem, we present two complementary hard instance mining strategies, i.e., Language-relevant Channel Filter and Relative Hard Instance Construction. They encourage the network to exclude visual-distinguishable feature and to focus on easy-confused objects during the contrastive training. Extensive experiments on two benchmarks, i.e., A2D Sentences and J-HMDB Sentences, quantitatively demonstrate the state-of-the-arts performance of our method and qualitatively show the more accurate distinguishment between semantically similar objects over baselines.

arXiv:2109.13921

Click-through Rate Prediction with Auto-Quantized Contrastive Learning

Click-through rate (CTR) prediction becomes indispensable in ubiquitous web recommendation applications. Nevertheless, the current methods are struggling under the cold-start scenarios where the user interactions are extremely sparse. We consider this problem as an automatic identification about whether the user behaviors are rich enough to capture the interests for prediction, and propose an Auto-Quantized Contrastive Learning (AQCL) loss to regularize the model. Different from previous methods, AQCL explores both the instance-instance and the instance-cluster similarity to robustify the latent representation, and automatically reduces the information loss to the active users due to the quantization. The proposed framework is agnostic to different model architectures and can be trained in an end-to-end fashion. Extensive results show that it consistently improves the current state-of-the-art CTR models.

arXiv:2109.13899

A Contrastive Learning Approach to Auroral Identification and Classification

Unsupervised learning algorithms are beginning to achieve accuracies comparable to their supervised counterparts on benchmark computer vision tasks, but their utility for practical applications has not yet been demonstrated. In this work, we present a novel application of unsupervised learning to the task of auroral image classification. Specifically, we modify and adapt the Simple framework for Contrastive Learning of Representations (SimCLR) algorithm to learn representations of auroral images in a recently released auroral image dataset constructed using image data from Time History of Events and Macroscale Interactions during Substorms (THEMIS) all-sky imagers. We demonstrate that (a) simple linear classifiers fit to the learned representations of the images achieve state-of-the-art classification performance, improving the classification accuracy by almost 10 percentage points over the current benchmark; and (b) the learned representations naturally cluster into more clusters than exist manually assigned categories, suggesting that existing categorizations are overly coarse and may obscure important connections between auroral types, near-earth solar wind conditions, and geomagnetic disturbances at the earth's surface. Moreover, our model is much lighter than the previous benchmark on this dataset, requiring in the area of fewer than 25\% of the number of parameters. Our approach exceeds an established threshold for operational purposes, demonstrating readiness for deployment and utilization.

arXiv:2109.13499

Modelling Neighbor Relation in Joint Space-Time Graph for Video Correspondence Learning

This paper presents a self-supervised method for learning reliable visual correspondence from unlabeled videos. We formulate the correspondence as finding paths in a joint space-time graph, where nodes are grid patches sampled from frames, and are linked by two types of edges: (i) neighbor relations that determine the aggregation strength from intra-frame neighbors in space, and (ii) similarity relations that indicate the transition probability of inter-frame paths across time. Leveraging the cycle-consistency in videos, our contrastive learning objective discriminates dynamic objects from both their neighboring views and temporal views. Compared with prior works, our approach actively explores the neighbor relations of central instances to learn a latent association between center-neighbor pairs (e.g., "hand -- arm") across time, thus improving the instance discrimination. Without fine-tuning, our learned representation outperforms the state-of-the-art self-supervised methods on a variety of visual tasks including video object propagation, part propagation, and pose keypoint tracking. Our self-supervised method also surpasses some fully supervised algorithms designed for the specific tasks.

arXiv:2109.13425

The JHU submission to VoxSRC-21: Track 3

This technical report describes Johns Hopkins University speaker recognition system submitted to Voxceleb Speaker Recognition Challenge 2021 Track 3: Self-supervised speaker verification (closed). Our overall training process is similar to the proposed one from the first place team in the last year's VoxSRC2020 challenge. The main difference is a recently proposed non-contrastive self-supervised method in computer vision (CV), distillation with no labels (DINO), is used to train our initial model, which outperformed the last year's contrastive learning based on momentum contrast (MoCo). Also, this requires only a few iterations in the iterative clustering stage, where pseudo labels for supervised embedding learning are updated based on the clusters of the embeddings generated from a model that is continually fine-tuned over iterations. In the final stage, Res2Net50 is trained on the final pseudo labels from the iterative clustering stage. This is our best submitted model to the challenge, showing 1.89, 6.50, and 6.89 in EER(%) in voxceleb1 test o, VoxSRC-21 validation, and test trials, respectively.

arXiv:2109.12599공식 코드 · Papers with Code

DialogueCSE: Dialogue-based Contrastive Learning of Sentence Embeddings

Learning sentence embeddings from dialogues has drawn increasing attention due to its low annotation cost and high domain adaptability. Conventional approaches employ the siamese-network for this task, which obtains the sentence embeddings through modeling the context-response semantic relevance by applying a feed-forward network on top of the sentence encoders. However, as the semantic textual similarity is commonly measured through the element-wise distance metrics (e.g. cosine and L2 distance), such architecture yields a large gap between training and evaluating. In this paper, we propose DialogueCSE, a dialogue-based contrastive learning approach to tackle this issue. DialogueCSE first introduces a novel matching-guided embedding (MGE) mechanism, which generates a context-aware embedding for each candidate response embedding (i.e. the context-free embedding) according to the guidance of the multi-turn context-response matching matrices. Then it pairs each context-aware embedding with its corresponding context-free embedding and finally minimizes the contrastive loss across all pairs. We evaluate our model on three multi-turn dialogue datasets: the Microsoft Dialogue Corpus, the Jing Dong Dialogue Corpus, and the E-commerce Dialogue Corpus. Evaluation results show that our approach significantly outperforms the baselines across all three datasets in terms of MAP and Spearman's correlation measures, demonstrating its effectiveness. Further quantitative experiments show that our approach achieves better performance when leveraging more dialogue context and remains robust when less training data is provided.

arXiv:2109.12242

Weakly Supervised Contrastive Learning for Chest X-Ray Report Generation

Radiology report generation aims at generating descriptive text from radiology images automatically, which may present an opportunity to improve radiology reporting and interpretation. A typical setting consists of training encoder-decoder models on image-report pairs with a cross entropy loss, which struggles to generate informative sentences for clinical diagnoses since normal findings dominate the datasets. To tackle this challenge and encourage more clinically-accurate text outputs, we propose a novel weakly supervised contrastive loss for medical report generation. Experimental results demonstrate that our method benefits from contrasting target reports with incorrect but semantically-close ones. It outperforms previous work on both clinical correctness and text generation metrics for two public benchmarks.

arXiv:2109.13059공식 코드 · Papers with Code

Trans-Encoder: Unsupervised sentence-pair modelling through self- and mutual-distillations

In NLP, a large volume of tasks involve pairwise comparison between two sequences (e.g. sentence similarity and paraphrase identification). Predominantly, two formulations are used for sentence-pair tasks: bi-encoders and cross-encoders. Bi-encoders produce fixed-dimensional sentence representations and are computationally efficient, however, they usually underperform cross-encoders. Cross-encoders can leverage their attention heads to exploit inter-sentence interactions for better performance but they require task fine-tuning and are computationally more expensive. In this paper, we present a completely unsupervised sentence representation model termed as Trans-Encoder that combines the two learning paradigms into an iterative joint framework to simultaneously learn enhanced bi- and cross-encoders. Specifically, on top of a pre-trained Language Model (PLM), we start with converting it to an unsupervised bi-encoder, and then alternate between the bi- and cross-encoder task formulations. In each alternation, one task formulation will produce pseudo-labels which are used as learning signals for the other task formulation. We then propose an extension to conduct such self-distillation approach on multiple PLMs in parallel and use the average of their pseudo-labels for mutual-distillation. Trans-Encoder creates, to the best of our knowledge, the first completely unsupervised cross-encoder and also a state-of-the-art unsupervised bi-encoder for sentence similarity. Both the bi-encoder and cross-encoder formulations of Trans-Encoder outperform recently proposed state-of-the-art unsupervised sentence encoders such as Mirror-BERT and SimCSE by up to 5% on the sentence similarity benchmarks.

arXiv:2109.12909공식 코드 · Papers with Code

Compressive Visual Representations

Learning effective visual representations that generalize well without human supervision is a fundamental problem in order to apply Machine Learning to a wide variety of tasks. Recently, two families of self-supervised methods, contrastive learning and latent bootstrapping, exemplified by SimCLR and BYOL respectively, have made significant progress. In this work, we hypothesize that adding explicit information compression to these algorithms yields better and more robust representations. We verify this by developing SimCLR and BYOL formulations compatible with the Conditional Entropy Bottleneck (CEB) objective, allowing us to both measure and control the amount of compression in the learned representation, and observe their impact on downstream tasks. Furthermore, we explore the relationship between Lipschitz continuity and compression, showing a tractable lower bound on the Lipschitz constant of the encoders we learn. As Lipschitz continuity is closely related to robustness, this provides a new explanation for why compressed models are more robust. Our experiments confirm that adding compression to SimCLR and BYOL significantly improves linear evaluation accuracies and model robustness across a wide range of domain shifts. In particular, the compressed version of BYOL achieves 76.0% Top-1 linear evaluation accuracy on ImageNet with ResNet-50, and 78.8% with ResNet-50 2x.

arXiv:2109.12714

Cluster Analysis with Deep Embeddings and Contrastive Learning

Unsupervised disentangled representation learning is a long-standing problem in computer vision. This work proposes a novel framework for performing image clustering from deep embeddings by combining instance-level contrastive learning with a deep embedding based cluster center predictor. Our approach jointly learns representations and predicts cluster centers in an end-to-end manner. This is accomplished via a three-pronged approach that combines a clustering loss, an instance-wise contrastive loss, and an anchor loss. Our fundamental intuition is that using an ensemble loss that incorporates instance-level features and a clustering procedure focusing on semantic similarity reinforces learning better representations in the latent space. We observe that our method performs exceptionally well on popular vision datasets when evaluated using standard clustering metrics such as Normalized Mutual Information (NMI), in addition to producing geometrically well-separated cluster embeddings as defined by the Euclidean distance. Our framework performs on par with widely accepted clustering methods and outperforms the state-of-the-art contrastive learning method on the CIFAR-10 dataset with an NMI score of 0.772, a 7-8% improvement on the strong baseline.

arXiv:2109.12493

Self-Supervised Video Representation Learning by Video Incoherence Detection

This paper introduces a novel self-supervised method that leverages incoherence detection for video representation learning. It roots from the observation that visual systems of human beings can easily identify video incoherence based on their comprehensive understanding of videos. Specifically, the training sample, denoted as the incoherent clip, is constructed by multiple sub-clips hierarchically sampled from the same raw video with various lengths of incoherence between each other. The network is trained to learn high-level representation by predicting the location and length of incoherence given the incoherent clip as input. Additionally, intra-video contrastive learning is introduced to maximize the mutual information between incoherent clips from the same raw video. We evaluate our proposed method through extensive experiments on action recognition and video retrieval utilizing various backbone networks. Experiments show that our proposed method achieves state-of-the-art performance across different backbone networks and different datasets compared with previous coherence-based methods.

arXiv:2109.12363

Contrastive Learning for Mitochondria Segmentation

Mitochondria segmentation in electron microscopy images is essential in neuroscience. However, due to the image degradation during the imaging process, the large variety of mitochondrial structures, as well as the presence of noise, artifacts and other sub-cellular structures, mitochondria segmentation is very challenging. In this paper, we propose a novel and effective contrastive learning framework to learn a better feature representation from hard examples to improve segmentation. Specifically, we adopt a point sampling strategy to pick out representative pixels from hard examples in the training phase. Based on these sampled pixels, we introduce a pixel-wise label-based contrastive loss which consists of a similarity loss term and a consistency loss term. The similarity term can increase the similarity of pixels from the same class and the separability of pixels from different classes in feature space, while the consistency term is able to enhance the sensitivity of the 3D model to changes in image content from frame to frame. We demonstrate the effectiveness of our method on MitoEM dataset as well as FIB-SEM dataset and show better or on par with state-of-the-art results.

arXiv:2109.12333공식 코드 · Papers with Code

Hard-sample Guided Hybrid Contrast Learning for Unsupervised Person Re-Identification

Unsupervised person re-identification (Re-ID) is a promising and very challenging research problem in computer vision. Learning robust and discriminative features with unlabeled data is of central importance to Re-ID. Recently, more attention has been paid to unsupervised Re-ID algorithms based on clustered pseudo-label. However, the previous approaches did not fully exploit information of hard samples, simply using cluster centroid or all instances for contrastive learning. In this paper, we propose a Hard-sample Guided Hybrid Contrast Learning (HHCL) approach combining cluster-level loss with instance-level loss for unsupervised person Re-ID. Our approach applies cluster centroid contrastive loss to ensure that the network is updated in a more stable way. Meanwhile, introduction of a hard instance contrastive loss further mines the discriminative information. Extensive experiments on two popular large-scale Re-ID benchmarks demonstrate that our HHCL outperforms previous state-of-the-art methods and significantly improves the performance of unsupervised person Re-ID. The code of our work is available soon at https://github.com/bupt-ai-cz/HHCL-ReID.

arXiv:2109.11730공식 코드 · Papers with Code

GeomGCL: Geometric Graph Contrastive Learning for Molecular Property Prediction

Recently many efforts have been devoted to applying graph neural networks (GNNs) to molecular property prediction which is a fundamental task for computational drug and material discovery. One of major obstacles to hinder the successful prediction of molecule property by GNNs is the scarcity of labeled data. Though graph contrastive learning (GCL) methods have achieved extraordinary performance with insufficient labeled data, most focused on designing data augmentation schemes for general graphs. However, the fundamental property of a molecule could be altered with the augmentation method (like random perturbation) on molecular graphs. Whereas, the critical geometric information of molecules remains rarely explored under the current GNN and GCL architectures. To this end, we propose a novel graph contrastive learning method utilizing the geometry of the molecule across 2D and 3D views, which is named GeomGCL. Specifically, we first devise a dual-view geometric message passing network (GeomMPNN) to adaptively leverage the rich information of both 2D and 3D graphs of a molecule. The incorporation of geometric properties at different levels can greatly facilitate the molecular representation learning. Then a novel geometric graph contrastive scheme is designed to make both geometric views collaboratively supervise each other to improve the generalization ability of GeomMPNN. We evaluate GeomGCL on various downstream property prediction tasks via a finetune process. Experimental results on seven real-life molecular datasets demonstrate the effectiveness of our proposed GeomGCL against state-of-the-art baselines.

arXiv:2109.11778

Dense Contrastive Visual-Linguistic Pretraining

Inspired by the success of BERT, several multimodal representation learning approaches have been proposed that jointly represent image and text. These approaches achieve superior performance by capturing high-level semantic information from large-scale multimodal pretraining. In particular, LXMERT and UNITER adopt visual region feature regression and label classification as pretext tasks. However, they tend to suffer from the problems of noisy labels and sparse semantic annotations, based on the visual features having been pretrained on a crowdsourced dataset with limited and inconsistent semantic labeling. To overcome these issues, we propose unbiased Dense Contrastive Visual-Linguistic Pretraining (DCVLP), which replaces the region regression and classification with cross-modality region contrastive learning that requires no annotations. Two data augmentation strategies (Mask Perturbation and Intra-/Inter-Adversarial Perturbation) are developed to improve the quality of negative samples used in contrastive learning. Overall, DCVLP allows cross-modality dense region contrastive learning in a self-supervised setting independent of any object annotations. We compare our method against prior visual-linguistic pretraining frameworks to validate the superiority of dense contrastive learning on multimodal representation learning.

arXiv:2109.11593

Long Short View Feature Decomposition via Contrastive Video Representation Learning

Self-supervised video representation methods typically focus on the representation of temporal attributes in videos. However, the role of stationary versus non-stationary attributes is less explored: Stationary features, which remain similar throughout the video, enable the prediction of video-level action classes. Non-stationary features, which represent temporally varying attributes, are more beneficial for downstream tasks involving more fine-grained temporal understanding, such as action segmentation. We argue that a single representation to capture both types of features is sub-optimal, and propose to decompose the representation space into stationary and non-stationary features via contrastive learning from long and short views, i.e. long video sequences and their shorter sub-sequences. Stationary features are shared between the short and long views, while non-stationary features aggregate the short views to match the corresponding long view. To empirically verify our approach, we demonstrate that our stationary features work particularly well on an action recognition downstream task, while our non-stationary features perform better on action segmentation. Furthermore, we analyse the learned representations and find that stationary features capture more temporally stable, static attributes, while non-stationary features encompass more temporally varying ones.

arXiv:2109.11475

Self-supervised Learning for Semi-supervised Temporal Language Grounding

Given a text description, Temporal Language Grounding (TLG) aims to localize temporal boundaries of the segments that contain the specified semantics in an untrimmed video. TLG is inherently a challenging task, as it requires comprehensive understanding of both sentence semantics and video contents. Previous works either tackle this task in a fully-supervised setting that requires a large amount of temporal annotations or in a weakly-supervised setting that usually cannot achieve satisfactory performance. Since manual annotations are expensive, to cope with limited annotations, we tackle TLG in a semi-supervised way by incorporating self-supervised learning, and propose Self-Supervised Semi-Supervised Temporal Language Grounding (S^4TLG). S^4TLG consists of two parts: (1) A pseudo label generation module that adaptively produces instant pseudo labels for unlabeled samples based on predictions from a teacher model; (2) A self-supervised feature learning module with inter-modal and intra-modal contrastive losses to learn video feature representations under the constraints of video content consistency and video-text alignment. We conduct extensive experiments on the ActivityNet-CD-OOD and Charades-CD-OOD datasets. The results demonstrate that our proposed S^4TLG can achieve competitive performance compared to fully-supervised state-of-the-art methods while only requiring a small portion of temporal annotations.

arXiv:2109.10967

Learning Contrastive Representation for Semantic Correspondence

Dense correspondence across semantically related images has been extensively studied, but still faces two challenges: 1) large variations in appearance, scale and pose exist even for objects from the same category, and 2) labeling pixel-level dense correspondences is labor intensive and infeasible to scale. Most existing approaches focus on designing various matching approaches with fully-supervised ImageNet pretrained networks. On the other hand, while a variety of self-supervised approaches are proposed to explicitly measure image-level similarities, correspondence matching the pixel level remains under-explored. In this work, we propose a multi-level contrastive learning approach for semantic matching, which does not rely on any ImageNet pretrained model. We show that image-level contrastive learning is a key component to encourage the convolutional features to find correspondence between similar objects, while the performance can be further enhanced by regularizing cross-instance cycle-consistency at intermediate feature levels. Experimental results on the PF-PASCAL, PF-WILLOW, and SPair-71k benchmark datasets demonstrate that our method performs favorably against the state-of-the-art approaches. The source code and trained models will be made available to the public.