Hash representation learning of multi-view heterogeneous data is the key to improving the accuracy of multimedia retrieval. However, existing methods utilize local similarity and fall short of deeply fusing the multi-view features, resulting in poor retrieval accuracy. Current methods only use local similarity to train their model. These methods ignore global similarity. Furthermore, most recent works fuse the multi-view features via a weighted sum or concatenation. We contend that these fusion methods are insufficient for capturing the interaction between various views. We present a novel Central Similarity Multi-View Hashing (CSMVH) method to address the mentioned problems. Central similarity learning is used for solving the local similarity problem, which can utilize the global similarity between the hash center and samples. We present copious empirical data demonstrating the superiority of gate-based fusion over conventional approaches. On the MS COCO and NUS-WIDE, the proposed CSMVH performs better than the state-of-the-art methods by a large margin (up to 11.41% mean Average Precision (mAP) improvement).
During the past decade, deep neural networks have led to fast-paced progress and significant achievements in computer vision problems, for both academia and industry. Yet despite their success, state-of-the-art image classification approaches fail to generalize well in previously unseen visual contexts, as required by many real-world applications. In this paper, we focus on this domain generalization (DG) problem and argue that the generalization ability of deep convolutional neural networks can be improved by taking advantage of multi-layer and multi-scaled representations of the network. We introduce a framework that aims at improving domain generalization of image classifiers by combining both low-level and high-level features at multiple scales, enabling the network to implicitly disentangle representations in its latent space and learn domain-invariant attributes of the depicted objects. Additionally, to further facilitate robust representation learning, we propose a novel objective function, inspired by contrastive learning, which aims at constraining the extracted representations to remain invariant under distribution shifts. We demonstrate the effectiveness of our method by evaluating on the domain generalization datasets of PACS, VLCS, Office-Home and NICO. Through extensive experimentation, we show that our model is able to surpass the performance of previous DG methods and consistently produce competitive and state-of-the-art results in all datasets
The development of transformer-based models has resulted in significant advances in addressing various vision and NLP-based research challenges. However, the progress made in transformer-based methods has not been effectively applied to biosensing research. This paper presents a novel Unified Biosensor-Vision Multi-modal Transformer-based (UBVMT) method to classify emotions in an arousal-valence space by combining a 2D representation of an ECG/PPG signal with the face information. To achieve this goal, we first investigate and compare the unimodal emotion recognition performance of three image-based representations of the ECG/PPG signal. We then present our UBVMT network which is trained to perform emotion recognition by combining the 2D image-based representation of the ECG/PPG signal and the facial expression features. Our unified transformer model consists of homogeneous transformer blocks that take as an input the 2D representation of the ECG/PPG signal and the corresponding face frame for emotion representation learning with minimal modality-specific design. Our UBVMT model is trained by reconstructing masked patches of video frames and 2D images of ECG/PPG signals, and contrastive modeling to align face and ECG/PPG data. Extensive experiments on the MAHNOB-HCI and DEAP datasets show that our Unified UBVMT-based model produces comparable results to the state-of-the-art techniques.
Endoscopy is a widely used technique for the early detection of diseases or robotic-assisted minimally invasive surgery (RMIS). Numerous deep learning (DL)-based research works have been developed for automated diagnosis or processing of endoscopic view. However, existing DL models may suffer from catastrophic forgetting. When new target classes are introduced over time or cross institutions, the performance of old classes may suffer severe degradation. More seriously, data privacy and storage issues may lead to the unavailability of old data when updating the model. Therefore, it is necessary to develop a continual learning (CL) methodology to solve the problem of catastrophic forgetting in endoscopic image segmentation. To tackle this, we propose a Endoscopy Continual Semantic Segmentation (EndoCSS) framework that does not involve the storage and privacy issues of exemplar data. The framework includes a mini-batch pseudo-replay (MB-PR) mechanism and a self-adaptive noisy cross-entropy (SAN-CE) loss. The MB-PR strategy circumvents privacy and storage issues by generating pseudo-replay images through a generative model. Meanwhile, the MB-PR strategy can also correct the model deviation to the replay data and current training data, which is aroused by the significant difference in the amount of current and replay images. Therefore, the model can perform effective representation learning on both new and old tasks. SAN-CE loss can help model fitting by adjusting the model's output logits, and also improve the robustness of training. Extensive continual semantic segmentation (CSS) experiments on public datasets demonstrate that our method can robustly and effectively address the catastrophic forgetting brought by class increment in endoscopy scenes. The results show that our framework holds excellent potential for real-world deployment in a streaming learning manner.
Model Checking is widely applied in verifying the correctness of complex and concurrent systems against a specification. Pure symbolic approaches while popular, suffer from the state space explosion problem due to cross product operations required that make them prohibitively expensive for large-scale systems and/or specifications. In this paper, we propose to use graph representation learning (GRL) for solving linear temporal logic (LTL) model checking, where the system and the specification are expressed by a B{\"u}chi automaton and an LTL formula, respectively. A novel GRL-based framework \model, is designed to learn the representation of the graph-structured system and specification, which reduces the model checking problem to binary classification. Empirical experiments on two model checking scenarios show that \model achieves promising accuracy, with up to $11\times$ overall speedup against canonical SOTA model checkers and $31\times$ for satisfiability checking alone.
Whilst contrastive learning yields powerful representations by matching different augmented views of the same instance, it lacks the ability to capture the similarities between different instances. One popular way to address this limitation is by learning global features (after the global pooling) to capture inter-instance relationships based on knowledge distillation, where the global features of the teacher are used to guide the learning of the global features of the student. Inspired by cross-modality learning, we extend this existing framework that only learns from global features by encouraging the global features and intermediate layer features to learn from each other. This leads to our novel self-supervised framework: cross-context learning between global and hypercolumn features (CGH), that enforces the consistency of instance relations between low- and high-level semantics. Specifically, we stack the intermediate feature maps to construct a hypercolumn representation so that we can measure instance relations using two contexts (hypercolumn and global feature) separately, and then use the relations of one context to guide the learning of the other. This cross-context learning allows the model to learn from the differences between the two contexts. The experimental results on linear classification and downstream tasks show that our method outperforms the state-of-the-art methods.
The atmosphere affects humans in a multitude of ways, from loss of life due to adverse weather effects to long-term social and economic impacts on societies. Computer simulations of atmospheric dynamics are, therefore, of great importance for the well-being of our and future generations. Here, we propose AtmoRep, a novel, task-independent stochastic computer model of atmospheric dynamics that can provide skillful results for a wide range of applications. AtmoRep uses large-scale representation learning from artificial intelligence to determine a general description of the highly complex, stochastic dynamics of the atmosphere from the best available estimate of the system's historical trajectory as constrained by observations. This is enabled by a novel self-supervised learning objective and a unique ensemble that samples from the stochastic model with a variability informed by the one in the historical record. The task-independent nature of AtmoRep enables skillful results for a diverse set of applications without specifically training for them and we demonstrate this for nowcasting, temporal interpolation, model correction, and counterfactuals. We also show that AtmoRep can be improved with additional data, for example radar observations, and that it can be extended to tasks such as downscaling. Our work establishes that large-scale neural networks can provide skillful, task-independent models of atmospheric dynamics. With this, they provide a novel means to make the large record of atmospheric observations accessible for applications and for scientific inquiry, complementing existing simulations based on first principles.
Tracking any given object(s) spatially and temporally is a common purpose in Visual Object Tracking (VOT) and Video Object Segmentation (VOS). Joint tracking and segmentation have been attempted in some studies but they often lack full compatibility of both box and mask in initialization and prediction, and mainly focus on single-object scenarios. To address these limitations, this paper proposes a Multi-object Mask-box Integrated framework for unified Tracking and Segmentation, dubbed MITS. Firstly, the unified identification module is proposed to support both box and mask reference for initialization, where detailed object information is inferred from boxes or directly retained from masks. Additionally, a novel pinpoint box predictor is proposed for accurate multi-object box prediction, facilitating target-oriented representation learning. All target objects are processed simultaneously from encoding to propagation and decoding, as a unified pipeline for VOT and VOS. Experimental results show MITS achieves state-of-the-art performance on both VOT and VOS benchmarks. Notably, MITS surpasses the best prior VOT competitor by around 6% on the GOT-10k test set, and significantly improves the performance of box initialization on VOS benchmarks. The code is available at https://github.com/yoxu515/MITS.
Hotspot detection using thermal imaging has recently become essential in several industrial applications, such as security applications, health applications, and equipment monitoring applications. Hotspot detection is of utmost importance in industrial safety where equipment can develop anomalies. Hotspots are early indicators of such anomalies. We address the problem of hotspot detection in thermal images by proposing a self-supervised learning approach. Self-supervised learning has shown potential as a competitive alternative to their supervised learning counterparts but their application to thermography has been limited. This has been due to lack of diverse data availability, domain specific pre-trained models, standardized benchmarks, etc. We propose a self-supervised representation learning approach followed by fine-tuning that improves detection of hotspots by classification. The SimSiam network based ensemble classifier decides whether an image contains hotspots or not. Detection of hotspots is followed by precise hotspot isolation. By doing so, we are able to provide a highly accurate and precise hotspot identification, applicable to a wide range of applications. We created a novel large thermal image dataset to address the issue of paucity of easily accessible thermal images. Our experiments with the dataset created by us and a publicly available segmentation dataset show the potential of our approach for hotspot detection and its ability to isolate hotspots with high accuracy. We achieve a Dice Coefficient of 0.736, the highest when compared with existing hotspot identification techniques. Our experiments also show self-supervised learning as a strong contender of supervised learning, providing competitive metrics for hotspot detection, with the highest accuracy of our approach being 97%.
While most research into speech synthesis has focused on synthesizing high-quality speech for in-dataset speakers, an equally essential yet unsolved problem is synthesizing speech for unseen speakers who are out-of-dataset with limited reference data, i.e., speaker adaptive speech synthesis. Many studies have proposed zero-shot speaker adaptive text-to-speech and voice conversion approaches aimed at this task. However, most current approaches suffer from the degradation of naturalness and speaker similarity when synthesizing speech for unseen speakers (i.e., speakers not in the training dataset) due to the poor generalizability of the model in out-of-distribution data. To address this problem, we propose GZS-TV, a generalizable zero-shot speaker adaptive text-to-speech and voice conversion model. GZS-TV introduces disentangled representation learning for both speaker embedding extraction and timbre transformation to improve model generalization and leverages the representation learning capability of the variational autoencoder to enhance the speaker encoder. Our experiments demonstrate that GZS-TV reduces performance degradation on unseen speakers and outperforms all baseline models in multiple datasets.
We address the problem of unsupervised semantic segmentation of outdoor LiDAR point clouds in diverse traffic scenarios. The key idea is to leverage the spatiotemporal nature of a dynamic point cloud sequence and introduce drastically stronger augmentation by establishing spatiotemporal correspondences across multiple frames. We dovetail clustering and pseudo-label learning in this work. Essentially, we alternate between clustering points into semantic groups and optimizing models using point-wise pseudo-spatiotemporal labels with a simple learning objective. Therefore, our method can learn discriminative features in an unsupervised learning fashion. We show promising segmentation performance on Semantic-KITTI, SemanticPOSS, and FLORIDA benchmark datasets covering scenarios in autonomous vehicle and intersection infrastructure, which is competitive when compared against many existing fully supervised learning methods. This general framework can lead to a unified representation learning approach for LiDAR point clouds incorporating domain knowledge.
Several recent works have directly extended the image masked autoencoder (MAE) with random masking into video domain, achieving promising results. However, unlike images, both spatial and temporal information are important for video understanding. This suggests that the random masking strategy that is inherited from the image MAE is less effective for video MAE. This motivates the design of a novel masking algorithm that can more efficiently make use of video saliency. Specifically, we propose a motion-guided masking algorithm (MGM) which leverages motion vectors to guide the position of each mask over time. Crucially, these motion-based correspondences can be directly obtained from information stored in the compressed format of the video, which makes our method efficient and scalable. On two challenging large-scale video benchmarks (Kinetics-400 and Something-Something V2), we equip video MAE with our MGM and achieve up to +$1.3\%$ improvement compared to previous state-of-the-art methods. Additionally, our MGM achieves equivalent performance to previous video MAE using up to $66\%$ fewer training epochs. Lastly, we show that MGM generalizes better to downstream transfer learning and domain adaptation tasks on the UCF101, HMDB51, and Diving48 datasets, achieving up to +$4.9\%$ improvement compared to baseline methods.
Self-supervised learning (SSL) techniques have recently produced outstanding results in learning visual representations from unlabeled videos. Despite the importance of motion in supervised learning techniques for action recognition, SSL methods often do not explicitly consider motion information in videos. To address this issue, we propose MOFO (MOtion FOcused), a novel SSL method for focusing representation learning on the motion area of a video, for action recognition. MOFO automatically detects motion areas in videos and uses these to guide the self-supervision task. We use a masked autoencoder which randomly masks out a high proportion of the input sequence; we force a specified percentage of the inside of the motion area to be masked and the remainder from outside. We further incorporate motion information into the finetuning step to emphasise motion in the downstream task. We demonstrate that our motion-focused innovations can significantly boost the performance of the currently leading SSL method (VideoMAE) for action recognition. Our method improves the recent self-supervised Vision Transformer (ViT), VideoMAE, by achieving +2.6%, +2.1%, +1.3% accuracy on Epic-Kitchens verb, noun and action classification, respectively, and +4.7% accuracy on Something-Something V2 action classification. Our proposed approach significantly improves the performance of the current SSL method for action recognition, indicating the importance of explicitly encoding motion in SSL.
As AI systems have obtained significant performance to be deployed widely in our daily live and human society, people both enjoy the benefits brought by these technologies and suffer many social issues induced by these systems. To make AI systems good enough and trustworthy, plenty of researches have been done to build guidelines for trustworthy AI systems. Machine learning is one of the most important parts for AI systems and representation learning is the fundamental technology in machine learning. How to make the representation learning trustworthy in real-world application, e.g., cross domain scenarios, is very valuable and necessary for both machine learning and AI system fields. Inspired by the concepts in trustworthy AI, we proposed the first trustworthy representation learning across domains framework which includes four concepts, i.e, robustness, privacy, fairness, and explainability, to give a comprehensive literature review on this research direction. Specifically, we first introduce the details of the proposed trustworthy framework for representation learning across domains. Second, we provide basic notions and comprehensively summarize existing methods for the trustworthy framework from four concepts. Finally, we conclude this survey with insights and discussions on future research directions.
In this work, we focus on robust time series representation learning. Our assumption is that real-world time series is noisy and complementary information from different views of the same time series plays an important role while analyzing noisy input. Based on this, we create two views for the input time series through two different encoders. We conduct co-training based contrastive learning iteratively to learn the encoders. Our experiments demonstrate that this co-training approach leads to a significant improvement in performance. Especially, by leveraging the complementary information from different views, our proposed TS-CoT method can mitigate the impact of data noise and corruption. Empirical evaluations on four time series benchmarks in unsupervised and semi-supervised settings reveal that TS-CoT outperforms existing methods. Furthermore, the representations learned by TS-CoT can transfer well to downstream tasks through fine-tuning.
Resting-state functional MRI (rs-fMRI) is increasingly employed in multi-site research to aid neurological disorder analysis. Existing studies usually suffer from significant cross-site/domain data heterogeneity caused by site effects such as differences in scanners/protocols. Many methods have been proposed to reduce fMRI heterogeneity between source and target domains, heavily relying on the availability of source data. But acquiring source data is challenging due to privacy concerns and/or data storage burdens in multi-site studies. To this end, we design a source-free collaborative domain adaptation (SCDA) framework for fMRI analysis, where only a pretrained source model and unlabeled target data are accessible. Specifically, a multi-perspective feature enrichment method (MFE) is developed for target fMRI analysis, consisting of multiple collaborative branches to dynamically capture fMRI features of unlabeled target data from multiple views. Each branch has a data-feeding module, a spatiotemporal feature encoder, and a class predictor. A mutual-consistency constraint is designed to encourage pair-wise consistency of latent features of the same input generated from these branches for robust representation learning. To facilitate efficient cross-domain knowledge transfer without source data, we initialize MFE using parameters of a pretrained source model. We also introduce an unsupervised pretraining strategy using 3,806 unlabeled fMRIs from three large-scale auxiliary databases, aiming to obtain a general feature encoder. Experimental results on three public datasets and one private dataset demonstrate the efficacy of our method in cross-scanner and cross-study prediction tasks. The model pretrained on large-scale rs-fMRI data has been released to the public.
Sound events in daily life carry rich information about the objective world. The composition of these sounds affects the mood of people in a soundscape. Most previous approaches only focus on classifying and detecting audio events and scenes, but may ignore their perceptual quality that may impact humans' listening mood for the environment, e.g. annoyance. To this end, this paper proposes a novel hierarchical graph representation learning (HGRL) approach which links objective audio events (AE) with subjective annoyance ratings (AR) of the soundscape perceived by humans. The hierarchical graph consists of fine-grained event (fAE) embeddings with single-class event semantics, coarse-grained event (cAE) embeddings with multi-class event semantics, and AR embeddings. Experiments show the proposed HGRL successfully integrates AE with AR for AEC and ARP tasks, while coordinating the relations between cAE and fAE and further aligning the two different grains of AE information with the AR.
Graph Convolutional Networks (GCNs) can capture non-Euclidean spatial dependence between different brain regions. The graph pooling operator, a crucial element of GCNs, enhances the representation learning capability and facilitates the acquisition of abnormal brain maps. However, most existing research designs graph pooling operators solely from the perspective of nodes while disregarding the original edge features. This confines graph pooling application scenarios and diminishes its ability to capture critical substructures. In this paper, we propose a novel edge-aware hard clustering graph pool (EHCPool), which is tailored to dominant edge features and redefines the clustering process. EHCPool initially introduced the 'Edge-to-Node' score criterion which utilized edge information to evaluate the significance of nodes. An innovative Iteration n-top strategy was then developed, guided by edge scores, to adaptively learn sparse hard clustering assignments for graphs. Additionally, a N-E Aggregation strategy is designed to aggregate node and edge features in each independent subgraph. Extensive experiments on the multi-site public datasets demonstrate the superiority and robustness of the proposed model. More notably, EHCPool has the potential to probe different types of dysfunctional brain networks from a data-driven perspective. Method code: https://github.com/swfen/EHCPool
We present a novel unsupervised domain adaption method for person re-identification (reID) that generalizes a model trained on a labeled source domain to an unlabeled target domain. We introduce a camera-driven curriculum learning (CaCL) framework that leverages camera labels of person images to transfer knowledge from source to target domains progressively. To this end, we divide target domain dataset into multiple subsets based on the camera labels, and initially train our model with a single subset (i.e., images captured by a single camera). We then gradually exploit more subsets for training, according to a curriculum sequence obtained with a camera-driven scheduling rule. The scheduler considers maximum mean discrepancies (MMD) between each subset and the source domain dataset, such that the subset closer to the source domain is exploited earlier within the curriculum. For each curriculum sequence, we generate pseudo labels of person images in a target domain to train a reID model in a supervised way. We have observed that the pseudo labels are highly biased toward cameras, suggesting that person images obtained from the same camera are likely to have the same pseudo labels, even for different IDs. To address the camera bias problem, we also introduce a camera-diversity (CD) loss encouraging person images of the same pseudo label, but captured across various cameras, to involve more for discriminative feature learning, providing person representations robust to inter-camera variations. Experimental results on standard benchmarks, including real-to-real and synthetic-to-real scenarios, demonstrate the effectiveness of our framework.
Sequential recommendation involves automatically recommending the next item to users based on their historical item sequence. While most prior research employs RNN or transformer methods to glean information from the item sequence-generating probabilities for each user-item pair and recommending the top items, these approaches often overlook the challenge posed by spurious relationships. This paper specifically addresses these spurious relations. We introduce a novel sequential recommendation framework named Irl4Rec. This framework harnesses invariant learning and employs a new objective that factors in the relationship between spurious variables and adjustment variables during model training. This approach aids in identifying spurious relations. Comparative analyses reveal that our framework outperforms three typical methods, underscoring the effectiveness of our model. Moreover, an ablation study further demonstrates the critical role our model plays in detecting spurious relations.