Gating modules have been widely explored in dynamic network pruning to reduce the run-time computational cost of deep neural networks while preserving the representation of features. Despite the substantial progress, existing methods remain ignoring the consistency between feature and gate distributions, which may lead to distortion of gated features. In this paper, we propose a feature-gate coupling (FGC) approach aiming to align distributions of features and gates. FGC is a plug-and-play module, which consists of two steps carried out in an iterative self-supervised manner. In the first step, FGC utilizes the $k$-Nearest Neighbor method in the feature space to explore instance neighborhood relationships, which are treated as self-supervisory signals. In the second step, FGC exploits contrastive learning to regularize gating modules with generated self-supervisory signals, leading to the alignment of instance neighborhood relationships within the feature and gate spaces. Experimental results validate that the proposed FGC method improves the baseline approach with significant margins, outperforming the state-of-the-arts with better accuracy-computation trade-off. Code is publicly available.
Multi-task learning (MTL) is an active field in deep learning in which we train a model to jointly learn multiple tasks by exploiting relationships between the tasks. It has been shown that MTL helps the model share the learned features between tasks and enhance predictions compared to when learning each task independently. We propose a new learning framework for 2-task MTL problem that uses the predictions of one task as inputs to another network to predict the other task. We define two new loss terms inspired by cycle-consistency loss and contrastive learning, alignment loss and cross-task consistency loss. Both losses are designed to enforce the model to align the predictions of multiple tasks so that the model predicts consistently. We theoretically prove that both losses help the model learn more efficiently and that cross-task consistency loss is better in terms of alignment with the straight-forward predictions. Experimental results also show that our proposed model achieves significant performance on the benchmark Cityscapes and NYU dataset.
Contrastive representation learning has proven to be an effective self-supervised learning method. Most successful approaches are based on Noise Contrastive Estimation (NCE) and use different views of an instance as positives that should be contrasted with other instances, called negatives, that are considered as noise. However, several instances in a dataset are drawn from the same distribution and share underlying semantic information. A good data representation should contain relations, or semantic similarity, between the instances. Contrastive learning implicitly learns relations but considering all negatives as noise harms the quality of the learned relations. To circumvent this issue, we propose a novel formulation of contrastive learning using semantic similarity between instances called Similarity Contrastive Estimation (SCE). Our training objective is a soft contrastive learning one. Instead of hard classifying positives and negatives, we estimate from one view of a batch a continuous distribution to push or pull instances based on their semantic similarities. This target similarity distribution is sharpened to eliminate noisy relations. The model predicts for each instance, from another view, the target distribution while contrasting its positive with negatives. Experimental results show that SCE is Top-1 on the ImageNet linear evaluation protocol at 100 pretraining epochs with 72.1% accuracy and is competitive with state-of-the-art algorithms by reaching 75.4% for 200 epochs with multi-crop. We also show that SCE is able to generalize to several tasks. Source code is available here: https://github.com/CEA-LIST/SCE.
Recent text-to-image matching models apply contrastive learning to large corpora of uncurated pairs of images and sentences. While such models can provide a powerful score for matching and subsequent zero-shot tasks, they are not capable of generating caption given an image. In this work, we repurpose such models to generate a descriptive text given an image at inference time, without any further training or tuning steps. This is done by combining the visual-semantic model with a large language model, benefiting from the knowledge in both web-scale models. The resulting captions are much less restrictive than those obtained by supervised captioning methods. Moreover, as a zero-shot learning method, it is extremely flexible and we demonstrate its ability to perform image arithmetic in which the inputs can be either images or text, and the output is a sentence. This enables novel high-level vision capabilities such as comparing two images or solving visual analogy tests. Our code is available at: https://github.com/YoadTew/zero-shot-image-to-text.
Contrastive learning has led to substantial improvements in the quality of learned embedding representations for tasks such as image classification. However, a key drawback of existing contrastive augmentation methods is that they may lead to the modification of the image content which can yield undesired alterations of its semantics. This can affect the performance of the model on downstream tasks. Hence, in this paper, we ask whether we can augment image data in contrastive learning such that the task-relevant semantic content of an image is preserved. For this purpose, we propose to leverage saliency-based explanation methods to create content-preserving masked augmentations for contrastive learning. Our novel explanation-driven supervised contrastive learning (ExCon) methodology critically serves the dual goals of encouraging nearby image embeddings to have similar content and explanation. To quantify the impact of ExCon, we conduct experiments on the CIFAR-100 and the Tiny ImageNet datasets. We demonstrate that ExCon outperforms vanilla supervised contrastive learning in terms of classification, explanation quality, adversarial robustness as well as probabilistic calibration in the context of distributional shift.
The goal of contrastive learning based pre-training is to leverage large quantities of unlabeled data to produce a model that can be readily adapted downstream. Current approaches revolve around solving an image discrimination task: given an anchor image, an augmented counterpart of that image, and some other images, the model must produce representations such that the distance between the anchor and its counterpart is small, and the distances between the anchor and the other images are large. There are two significant problems with this approach: (i) by contrasting representations at the image-level, it is hard to generate detailed object-sensitive features that are beneficial to downstream object-level tasks such as instance segmentation; (ii) the augmentation strategy of producing an augmented counterpart is fixed, making learning less effective at the later stages of pre-training. In this work, we introduce Curricular Contrastive Object-level Pre-training (CCOP) to tackle these problems: (i) we use selective search to find rough object regions and use them to build an inter-image object-level contrastive loss and an intra-image object-level discrimination loss into our pre-training objective; (ii) we present a curriculum learning mechanism that adaptively augments the generated regions, which allows the model to consistently acquire a useful learning signal, even in the later stages of pre-training. Our experiments show that our approach improves on the MoCo v2 baseline by a large margin on multiple object-level tasks when pre-training on multi-object scene image datasets. Code is available at https://github.com/ChenhongyiYang/CCOP.
High annotation costs are a substantial bottleneck in applying modern deep learning architectures to clinically relevant medical use cases, substantiating the need for novel algorithms to learn from unlabeled data. In this work, we propose ContIG, a self-supervised method that can learn from large datasets of unlabeled medical images and genetic data. Our approach aligns images and several genetic modalities in the feature space using a contrastive loss. We design our method to integrate multiple modalities of each individual person in the same model end-to-end, even when the available modalities vary across individuals. Our procedure outperforms state-of-the-art self-supervised methods on all evaluated downstream benchmark tasks. We also adapt gradient-based explainability algorithms to better understand the learned cross-modal associations between the images and genetic modalities. Finally, we perform genome-wide association studies on the features learned by our models, uncovering interesting relationships between images and genetic data.
Although there is significant progress in supervised semantic segmentation, it remains challenging to deploy the segmentation models to unseen domains due to domain biases. Domain adaptation can help in this regard by transferring knowledge from a labeled source domain to an unlabeled target domain. Previous methods typically attempt to perform the adaptation on global features, however, the local semantic affiliations accounting for each pixel in the feature space are often ignored, resulting in less discriminability. To solve this issue, we propose a novel semantic prototype-based contrastive learning framework for fine-grained class alignment. Specifically, the semantic prototypes provide supervisory signals for per-pixel discriminative representation learning and each pixel of source and target domains in the feature space is required to reflect the content of the corresponding semantic prototype. In this way, our framework is able to explicitly make intra-class pixel representations closer and inter-class pixel representations further apart to improve the robustness of the segmentation model as well as alleviate the domain shift problem. Our method is easy to implement and attains superior results compared to state-of-the-art approaches, as is demonstrated with a number of experiments. The code is publicly available at https://github.com/BinhuiXie/SPCL.
Self-supervised methods (SSL) have achieved significant success via maximizing the mutual information between two augmented views, where cropping is a popular augmentation technique. Cropped regions are widely used to construct positive pairs, while the left regions after cropping have rarely been explored in existing methods, although they together constitute the same image instance and both contribute to the description of the category. In this paper, we make the first attempt to demonstrate the importance of both regions in cropping from a complete perspective and propose a simple yet effective pretext task called Region Contrastive Learning (RegionCL). Specifically, given two different images, we randomly crop a region (called the paste view) from each image with the same size and swap them to compose two new images together with the left regions (called the canvas view), respectively. Then, contrastive pairs can be efficiently constructed according to the following simple criteria, i.e., each view is (1) positive with views augmented from the same original image and (2) negative with views augmented from other images. With minor modifications to popular SSL methods, RegionCL exploits those abundant pairs and helps the model distinguish the regions features from both canvas and paste views, therefore learning better visual representations. Experiments on ImageNet, MS COCO, and Cityscapes demonstrate that RegionCL improves MoCo v2, DenseCL, and SimSiam by large margins and achieves state-of-the-art performance on classification, detection, and segmentation tasks. The code will be available at https://github.com/Annbless/RegionCL.git.
Recent self-supervised models have demonstrated equal or better performance than supervised methods, opening for AI systems to learn visual representations from practically unlimited data. However, these methods are typically classification-based and thus ineffective for learning high-resolution feature maps that preserve precise spatial information. This work introduces superpixels to improve self-supervised learning of dense semantically rich visual concept embeddings. Decomposing images into a small set of visually coherent regions reduces the computational complexity by $\mathcal{O}(1000)$ while preserving detail. We experimentally show that contrasting over regions improves the effectiveness of contrastive learning methods, extends their applicability to high-resolution images, improves overclustering performance, superpixels are better than grids, and regional masking improves performance. The expressiveness of our dense embeddings is demonstrated by improving the SOTA unsupervised semantic segmentation benchmark on Cityscapes, and for convolutional models on COCO.
Existing deep clustering methods rely on either contrastive or non-contrastive representation learning for downstream clustering task. Contrastive-based methods thanks to negative pairs learn uniform representations for clustering, in which negative pairs, however, may inevitably lead to the class collision issue and consequently compromise the clustering performance. Non-contrastive-based methods, on the other hand, avoid class collision issue, but the resulting non-uniform representations may cause the collapse of clustering. To enjoy the strengths of both worlds, this paper presents a novel end-to-end deep clustering method with prototype scattering and positive sampling, termed ProPos. Specifically, we first maximize the distance between prototypical representations, named prototype scattering loss, which improves the uniformity of representations. Second, we align one augmented view of instance with the sampled neighbors of another view -- assumed to be truly positive pair in the embedding space -- to improve the within-cluster compactness, termed positive sampling alignment. The strengths of ProPos are avoidable class collision issue, uniform representations, well-separated clusters, and within-cluster compactness. By optimizing ProPos in an end-to-end expectation-maximization framework, extensive experimental results demonstrate that ProPos achieves competing performance on moderate-scale clustering benchmark datasets and establishes new state-of-the-art performance on large-scale datasets. Source code is available at \url{https://github.com/Hzzone/ProPos}.
This paper presents a simple yet effective two-stage framework for semi-supervised medical image segmentation. Unlike prior state-of-the-art semi-supervised segmentation methods that predominantly rely on pseudo supervision directly on predictions, such as consistency regularization and pseudo labeling, our key insight is to explore the feature representation learning with labeled and unlabeled (i.e., pseudo labeled) images to regularize a more compact and better-separated feature space, which paves the way for low-density decision boundary learning and therefore enhances the segmentation performance. A stage-adaptive contrastive learning method is proposed, containing a boundary-aware contrastive loss that takes advantage of the labeled images in the first stage, as well as a prototype-aware contrastive loss to optimize both labeled and pseudo labeled images in the second stage. To obtain more accurate prototype estimation, which plays a critical role in prototype-aware contrastive learning, we present an aleatoric uncertainty-aware method, namely AUA, to generate higher-quality pseudo labels. AUA adaptively regularizes prediction consistency by taking advantage of image ambiguity, which, given its significance, is under-explored by existing works. Our method achieves the best results on three public medical image segmentation benchmarks.
Lesion detection is a fundamental problem in the computer-aided diagnosis scheme for mammography. The advance of deep learning techniques have made a remarkable progress for this task, provided that the training data are large and sufficiently diverse in terms of image style and quality. In particular, the diversity of image style may be majorly attributed to the vendor factor. However, the collection of mammograms from vendors as many as possible is very expensive and sometimes impractical for laboratory-scale studies. Accordingly, to further augment the generalization capability of deep learning model to various vendors with limited resources, a new contrastive learning scheme is developed. Specifically, the backbone network is firstly trained with a multi-style and multi-view unsupervised self-learning scheme for the embedding of invariant features to various vendor-styles. Afterward, the backbone network is then recalibrated to the downstream task of lesion detection with the specific supervised learning. The proposed method is evaluated with mammograms from four vendors and one unseen public dataset. The experimental results suggest that our approach can effectively improve detection performance on both seen and unseen domains, and outperforms many state-of-the-art (SOTA) generalization methods.
The ability to reliably estimate physiological signals from video is a powerful tool in low-cost, pre-clinical health monitoring. In this work we propose a new approach to remote photoplethysmography (rPPG) - the measurement of blood volume changes from observations of a person's face or skin. Similar to current state-of-the-art methods for rPPG, we apply neural networks to learn deep representations with invariance to nuisance image variation. In contrast to such methods, we employ a fully self-supervised training approach, which has no reliance on expensive ground truth physiological training data. Our proposed method uses contrastive learning with a weak prior over the frequency and temporal smoothness of the target signal of interest. We evaluate our approach on four rPPG datasets, showing that comparable or better results can be achieved compared to recent supervised deep learning methods but without using any annotation. In addition, we incorporate a learned saliency resampling module into both our unsupervised approach and supervised baseline. We show that by allowing the model to learn where to sample the input image, we can reduce the need for hand-engineered features while providing some interpretability into the model's behavior and possible failure modes. We release code for our complete training and evaluation pipeline to encourage reproducible progress in this exciting new direction.
The reconstruction of microbial genomes from large metagenomic datasets is a critical procedure for finding uncultivated microbial populations and defining their microbial functional roles. To achieve that, we need to perform metagenomic binning, clustering the assembled contigs into draft genomes. Despite the existing computational tools, most of them neglect one important property of the metagenomic data, that is, the noise. To further improve the metagenomic binning step and reconstruct better metagenomes, we propose a deep Contrastive Learning framework for Metagenome Binning (CLMB), which can efficiently eliminate the disturbance of noise and produce more stable and robust results. Essentially, instead of denoising the data explicitly, we add simulated noise to the training data and force the deep learning model to produce similar and stable representations for both the noise-free data and the distorted data. Consequently, the trained model will be robust to noise and handle it implicitly during usage. CLMB outperforms the previous state-of-the-art binning methods significantly, recovering the most near-complete genomes on almost all the benchmarking datasets (up to 17\% more reconstructed genomes compared to the second-best method). It also improves the performance of bin refinement, reconstructing 8-22 more high-quality genomes and 15-32 more middle-quality genomes than the second-best result. Impressively, in addition to being compatible with the binning refiner, single CLMB even recovers on average 15 more HQ genomes than the refiner of VAMB and Maxbin on the benchmarking datasets. CLMB is open-source and available at https://github.com/zpf0117b/CLMB/.
Deep learning algorithms have recently achieved promising deraining performances on both the natural and synthetic rainy datasets. As an essential low-level pre-processing stage, a deraining network should clear the rain streaks and preserve the fine semantic details. However, most existing methods only consider low-level image restoration. That limits their performances at high-level tasks requiring precise semantic information. To address this issue, in this paper, we present a segmentation-aware progressive network (SAPNet) based upon contrastive learning for single image deraining. We start our method with a lightweight derain network formed with progressive dilated units (PDU). The PDU can significantly expand the receptive field and characterize multi-scale rain streaks without the heavy computation on multi-scale images. A fundamental aspect of this work is an unsupervised background segmentation (UBS) network initialized with ImageNet and Gaussian weights. The UBS can faithfully preserve an image's semantic information and improve the generalization ability to unseen photos. Furthermore, we introduce a perceptual contrastive loss (PCL) and a learned perceptual image similarity loss (LPISL) to regulate model learning. By exploiting the rainy image and groundtruth as the negative and the positive sample in the VGG-16 latent space, we bridge the fine semantic details between the derained image and the groundtruth in a fully constrained manner. Comprehensive experiments on synthetic and real-world rainy images show our model surpasses top-performing methods and aids object detection and semantic segmentation with considerable efficacy. A Pytorch Implementation is available at https://github.com/ShenZheng2000/SAPNet-for-image-deraining.
Clustering of hyperspectral images is a fundamental but challenging task. The recent development of hyperspectral image clustering has evolved from shallow models to deep and achieved promising results in many benchmark datasets. However, their poor scalability, robustness, and generalization ability, mainly resulting from their offline clustering scenarios, greatly limit their application to large-scale hyperspectral data. To circumvent these problems, we present a scalable deep online clustering model, named Spectral-Spatial Contrastive Clustering (SSCC), based on self-supervised learning. Specifically, we exploit a symmetric twin neural network comprised of a projection head with a dimensionality of the cluster number to conduct dual contrastive learning from a spectral-spatial augmentation pool. We define the objective function by implicitly encouraging within-cluster similarity and reducing between-cluster redundancy. The resulting approach is trained in an end-to-end fashion by batch-wise optimization, making it robust in large-scale data and resulting in good generalization ability for unseen data. Extensive experiments on three hyperspectral image benchmarks demonstrate the effectiveness of our approach and show that we advance the state-of-the-art approaches by large margins.
Generative models have been applied in the medical imaging domain for various image recognition and synthesis tasks. However, a more controllable and interpretable image synthesis model is still lacking yet necessary for important applications such as assisting in medical training. In this work, we leverage the efficient self-attention and contrastive learning modules and build upon state-of-the-art generative adversarial networks (GANs) to achieve an attribute-aware image synthesis model, termed AttributeGAN, which can generate high-quality histopathology images based on multi-attribute inputs. In comparison to existing single-attribute conditional generative models, our proposed model better reflects input attributes and enables smoother interpolation among attribute values. We conduct experiments on a histopathology dataset containing stained H&E images of urothelial carcinoma and demonstrate the effectiveness of our proposed model via comprehensive quantitative and qualitative comparisons with state-of-the-art models as well as different variants of our model. Code is available at https://github.com/karenyyy/MICCAI2021AttributeGAN.
Contrastive learning can largely enhance the feature discriminability in a self-supervised manner and has achieved remarkable success for various visual tasks. However, it is undesirably observed that the standard contrastive paradigm (features+$\ell_{2}$ normalization) only brings little help for domain adaptation. In this work, we delve into this phenomenon and find that the main reason is due to the class weights (weights of the final fully connected layer) which are vital for the recognition yet ignored in the optimization. To tackle this issue, we propose a simple yet powerful Probabilistic Contrastive Learning (PCL), which does not only assist in extracting discriminative features but also enforces them to be clustered around the class weights. Specifically, we break the standard contrastive paradigm by removing $\ell_{2}$ normalization and replacing the features with probabilities. In this way, PCL can enforce the probability to approximate the one-hot form, thereby reducing the deviation between the features and class weights. Benefiting from the conciseness, PCL can be well generalized to different settings. In this work, we conduct extensive experiments on five tasks and observe consistent performance gains, i.e., Unsupervised Domain Adaptation (UDA), Semi-Supervised Domain Adaptation (SSDA), Semi-Supervised Learning (SSL), UDA Detection, and UDA Semantic Segmentation. Notably, for UDA Semantic Segmentation on SYNTHIA, PCL surpasses the sophisticated CPSL-D by $>\!2\%$ in terms of mean IoU with a much smaller training cost (PCL: 1*3090, 5 days v.s. CPSL-D: 4*V100, 11 days). Code is available at https://github.com/ljjcoder/Probabilistic-Contrastive-Learning.
Humans can easily segment moving objects without knowing what they are. That objectness could emerge from continuous visual observations motivates us to model grouping and movement concurrently from unlabeled videos. Our premise is that a video has different views of the same scene related by moving components, and the right region segmentation and region flow would allow mutual view synthesis which can be checked from the data itself without any external supervision. Our model starts with two separate pathways: an appearance pathway that outputs feature-based region segmentation for a single image, and a motion pathway that outputs motion features for a pair of images. It then binds them in a conjoint representation called segment flow that pools flow offsets over each region and provides a gross characterization of moving regions for the entire scene. By training the model to minimize view synthesis errors based on segment flow, our appearance and motion pathways learn region segmentation and flow estimation automatically without building them up from low-level edges or optical flows respectively. Our model demonstrates the surprising emergence of objectness in the appearance pathway, surpassing prior works on zero-shot object segmentation from an image, moving object segmentation from a video with unsupervised test-time adaptation, and semantic image segmentation by supervised fine-tuning. Our work is the first truly end-to-end zero-shot object segmentation from videos. It not only develops generic objectness for segmentation and tracking, but also outperforms prevalent image-based contrastive learning methods without augmentation engineering.