Explore the vast range of titles available.

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 methods focus on designing various matching modules using 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.

Therapeutic peptides contribute significantly to human health and have the potential for personalized medicine. The prediction for the therapeutic peptides is beneficial and emerging for the discovery of drugs. Although several computational approaches have emerged to discern the functions of therapeutic peptides, predicting multi-functional therapeutic peptide types is challenging. In this research, a novel approach termed TPpred-SC has been introduced. This method leverages a pretrained protein language model alongside multi-label supervised contrastive learning to predict multi-functional therapeutic peptides. The framework incorporates sequential semantic information directly from large-scale protein sequences in TAPE. Then, TPpred-SC exploits multi-label supervised contrastive learning to enhance the representation of peptide sequences for imbalanced multi-label therapeutic peptide prediction. The experimental findings demonstrate that TPpred-SC achieves superior performance compared to existing related methods. To serve our work more efficiently, the web server of TPpred-SC can be accessed at http://bliulab.net/TPpred-SC .

In this letter, we developed novel Structure Enhanced Channel Attention Network (SECANet) for scene text image super‐resolution (STISR). The newly proposed SECANet integrates a group of Structure‐Enhanced Attention Modules to focus more on both local and global structural features in the character regions of text images. Moreover, we elaborately formulate a Dual‐Domain Contrastive Learning framework that integrates one pixel‐level contrastive loss and the other semantic‐level contrastive loss to jointly optimize the SECANet for generating more visually pleasing yet better recognizable high‐quality SR images without introducing any additional prior generators in both the training and testing stages, showing promising computational efficiency. Experimental results on the Textzoom dataset indicate that our method can achieve both decent performance in super‐resolving more impressive scene text images from low‐resolution ones and better recognition accuracy than other competitors. The proposed structure enhanced channel attention network assembles a group of structure‐enhanced attention blocks to learn both global and local structure features for the detailed recovery of scene text images. Moreover, a joint dual‐domain contrastive loss function is formulated to optimize the model parameters, benefiting to synthesizing more recognizable text images.

In this paper, we propose a physics-inspired contrastive learning paradigm for low-light enhancement, called PIE. PIE primarily addresses three issues: (i) To resolve the problem of existing learning-based methods often training a LLE model with strict pixel-correspondence image pairs, we eliminate the need for pixel-correspondence paired training data and instead train with unpaired images. (ii) To address the disregard for negative samples and the inadequacy of their generation in existing methods, we incorporate physics-inspired contrastive learning for LLE and design the Bag of Curves (BoC) method to generate more reasonable negative samples that closely adhere to the underlying physical imaging principle. (iii) To overcome the reliance on semantic ground truths in existing methods, we propose an unsupervised regional segmentation module, ensuring regional brightness consistency while eliminating the dependency on semantic ground truths. Overall, the proposed PIE can effectively learn from unpaired positive/negative samples and smoothly realize non-semantic regional enhancement, which is clearly different from existing LLE efforts. Besides the novel architecture of PIE, we explore the gain of PIE on downstream tasks such as semantic segmentation and face detection. Training on readily available open data and extensive experiments demonstrate that our method surpasses the state-of-the-art LLE models over six independent cross-scenes datasets. PIE runs fast with reasonable GFLOPs in test time, making it easy to use on mobile devices. Code available

With rapid advancements in depth sensors and deep learning, skeleton-based person re-identification (re-ID) models have recently achieved remarkable progress with many advantages. Most existing solutions learn single-level skeleton features from body joints with the assumption of equal skeleton importance, while they typically lack the ability to exploit more informative skeleton features from various levels such as limb level with more global body patterns. The label dependency of these methods also limits their flexibility in learning more general skeleton representations. This paper proposes a generic unsupervised Hierarchical skeleton Meta-Prototype Contrastive learning (Hi-MPC) approach with Hard Skeleton Mining (HSM) for person re-ID with unlabeled 3D skeletons. Firstly, we construct hierarchical representations of skeletons to model coarse-to-fine body and motion features from the levels of body joints, components, and limbs. Then a hierarchical meta-prototype contrastive learning model is proposed to cluster and contrast the most typical skeleton features (“prototypes”) from different-level skeletons. By converting original prototypes into meta-prototypes with multiple homogeneous transformations, we induce the model to learn the inherent consistency of prototypes to capture more effective skeleton features for person re-ID. Furthermore, we devise a hard skeleton mining mechanism to adaptively infer the informative importance of each skeleton, so as to focus on harder skeletons to learn more discriminative skeleton representations. Extensive evaluations on five datasets demonstrate that our approach outperforms a wide variety of state-of-the-art skeleton-based methods. We further show the general applicability of our method to cross-view person re-ID and RGB-based scenarios with estimated skeletons.

Deep neural networks have achieved significant success in the artificial intelligence community and various downstream tasks. They encode images or texts into dense feature representations and are supervised by a large amount of labeled data. Due to the expensiveness of high-quality labeled data, a huge number of easy-to-access instances are collected to conduct supervised learning. However, they have not been annotated by experts and thus can contain numerous noisy instances, which will degrade the performance. To learn robust feature representations despite misleading noisy labels, we employ supervised contrastive learning to directly perform supervision in the hidden space, rather than in the prediction space like the prevalent cross-entropy loss function. However, cutting-edge noisy label learning methods with supervised contrastive learning always discard the data considered to be noisy, and thus cannot tolerate high-ratio noisy datasets. Therefore, we propose a novel training strategy named Supervised Contrastive Learning with Corrected Labels (S cl 2 ) to defend against the attack of noisy labels. S cl 2 corrects the noisy labels with an empirical small-loss assumption and conducts supervised contrastive learning using these corrected data. Specifically, we employ the generated soft labels as supervisory information to facilitate our implementation of supervised contrastive learning. This expansion of contrastive learning ensures the integrity of the supervisory information while effectively enhancing the learning process. In addition, samples sharing the same soft labels are treated as positive sample pairs, while those with different soft labels are considered to be negative sample pairs. With this strategy, the representations from neural networks keep the local discrimination in one mini-batch. Besides, we also employ a prototype contrastive learning technique to ensure global discrimination. Our S cl 2 has demonstrated excellent performance on numerous benchmark datasets, showcasing its effectiveness in various standardized evaluation scenarios. Additionally, our model has proven to be highly valuable when applied to real-world noisy datasets.

Self-supervised graph representation learning has recently shown considerable promise in a range of fields, including bioinformatics and social networks. A large number of graph contrastive learning approaches have shown promising performance for representation learning on graphs, which train models by maximizing agreement between original graphs and their augmented views (i.e., positive views). Unfortunately, these methods usually involve pre-defined augmentation strategies based on the knowledge of human experts. Moreover, these strategies may fail to generate challenging positive views to provide sufficient supervision signals. In this paper, we present a novel approach named graph pooling contrast (GPS) to address these issues. Motivated by the fact that graph pooling can adaptively coarsen the graph with the removal of redundancy, we rethink graph pooling and leverage it to automatically generate multi-scale positive views with varying emphasis on providing challenging positives and preserving semantics, i.e., strongly-augmented view and weakly-augmented view. Then, we incorporate both views into a joint contrastive learning framework with similarity learning and consistency learning, where our pooling module is adversarially trained with respect to the encoder for adversarial robustness. Experiments on twelve datasets on both graph classification and transfer learning tasks verify the superiority of the proposed method over its counterparts.

The progress of EEG-based emotion recognition has received widespread attention from the fields of human-machine interaction and cognitive science. However, recognizing emotions with limited labelled data is still challenging. To address this issue, this paper proposes a self-supervised group meiosis contrastive learning (SGMC) framework for EEG-based emotion recognition. First, to reduce the dependence of emotion labels, SGMC introduces a contrastive learning task according to the alignment of video clips based on the similar EEG response across subjects. Moreover, the model adopts a group projector to extract group-level representations from the group samples to further decrease the subject difference and random effects in EEG signals. Finally, a novel genetics-inspired data augmentation method, named meiosis is developed, which takes advantage of the alignment of video clips among a group of EEG samples to generate augmented groups by pairing, cross exchanging, and separating. The experiments show that SGMC exhibits competitive performance on the publicly available DEAP and SEED datasets. It is worth of noting that the SGMC shows a high ability to recognize emotion even when using limited labelled data. Moreover, the results of feature visualization suggest that the model might have learned video-level emotion-related feature representations to improve emotion recognition. The hyper-parametric analysis further shows the effect of the group size during emotion recognition. Finally, the comparisons of both the symmetric function and the ablation models and the analysis of computational efficiency are carried out to examine the rationality of the SGMC architecture. The code is provided publicly online.

Diseases cause crop yield reduction and quality decline, which has a great impact on agricultural production. Plant disease recognition based on computer vision can help farmers quickly and accurately recognize diseases. However, the occurrence of diseases is random and the collection cost is very high. In many cases, the number of disease samples that can be used to train the disease classifier is small. To address this problem, we propose a few-shot disease recognition algorithm that uses supervised contrastive learning. Our algorithm is divided into two phases: supervised contrastive learning and meta-learning. In the first phase, we use a supervised contrastive learning algorithm to train an encoder with strong generalization capabilities using a large number of samples. In the second phase, we treat this encoder as an extractor of plant disease features and adopt the meta-learning training mechanism to accomplish the few-shot disease recognition tasks by training a nearest-centroid classifier based on distance metrics. The experimental results indicate that the proposed method outperforms the other nine popular few-shot learning algorithms as a comparison in the disease recognition accuracy over the public plant disease dataset PlantVillage. In few-shot potato leaf disease recognition tasks in natural scenarios, the accuracy of the model reaches the accuracy of 79.51% with only 30 training images. The experiment also revealed that, in the contrastive learning phase, the combination of different image augmentation operations has a greater impact on model. Furthermore, the introduction of label information in supervised contrastive learning enables our algorithm to still obtain high accuracy in few-shot disease recognition tasks with smaller batch size, thus allowing us to complete the training with less GPU resource compared to traditional contrastive learning.

Human Activity Recognition (HAR) using wearable sensors has shown great potential for personalized health management and ubiquitous computing. However, existing deep learning-based HAR models often suffer from poor user-level generalization, which limits their deployment in real-world scenarios. In this work, we propose a novel multi-task contrastive learning framework that jointly optimizes activity classification and supervised contrastive objectives to enhance generalization across unseen users. By leveraging both activity and user labels to construct semantically meaningful contrastive pairs, our method improves representation learning while maintaining user-agnostic inference at test time. We evaluate the proposed framework on three public HAR datasets using cross-user splits, achieving comparable results to both supervised and self-supervised baselines. Extensive ablation studies further confirm the effectiveness of our design choices, including multi-task training and the integration of user-aware contrastive supervision. These results highlight the potential of our approach for building more generalizable and scalable HAR systems.