In dialog system, dialog act recognition and sentiment classification are two correlative tasks to capture speakers intentions, where dialog act and sentiment can indicate the explicit and the implicit intentions separately. Most of the existing systems either treat them as separate tasks or just jointly model the two tasks by sharing parameters in an implicit way without explicitly modeling mutual interaction and relation. To address this problem, we propose a Deep Co-Interactive Relation Network (DCR-Net) to explicitly consider the cross-impact and model the interaction between the two tasks by introducing a co-interactive relation layer. In addition, the proposed relation layer can be stacked to gradually capture mutual knowledge with multiple steps of interaction. Especially, we thoroughly study different relation layers and their effects. Experimental results on two public datasets (Mastodon and Dailydialog) show that our model outperforms the state-of-the-art joint model by 4.3% and 3.4% in terms of F1 score on dialog act recognition task, 5.7% and 12.4% on sentiment classification respectively. Comprehensive analysis empirically verifies the effectiveness of explicitly modeling the relation between the two tasks and the multi-steps interaction mechanism. Finally, we employ the Bidirectional Encoder Representation from Transformer (BERT) in our framework, which can further boost our performance in both tasks.
Accurate and automated gland segmentation on histology tissue images is an essential but challenging task in the computer-aided diagnosis of adenocarcinoma. Despite their prevalence, deep learning models always require a myriad number of densely annotated training images, which are difficult to obtain due to extensive labor and associated expert costs related to histology image annotations. In this paper, we propose the pairwise relation-based semi-supervised (PRS^2) model for gland segmentation on histology images. This model consists of a segmentation network (S-Net) and a pairwise relation network (PR-Net). The S-Net is trained on labeled data for segmentation, and PR-Net is trained on both labeled and unlabeled data in an unsupervised way to enhance its image representation ability via exploiting the semantic consistency between each pair of images in the feature space. Since both networks share their encoders, the image representation ability learned by PR-Net can be transferred to S-Net to improve its segmentation performance. We also design the object-level Dice loss to address the issues caused by touching glands and combine it with other two loss functions for S-Net. We evaluated our model against five recent methods on the GlaS dataset and three recent methods on the CRAG dataset. Our results not only demonstrate the effectiveness of the proposed PR-Net and object-level Dice loss, but also indicate that our PRS^2 model achieves the state-of-the-art gland segmentation performance on both benchmarks.
Understanding and predicting pedestrian behavior is an important and challenging area of research for realizing safe and effective navigation strategies in automated and advanced driver assistance technologies in urban scenes. This paper focuses on monocular pedestrian action recognition and 3D localization from an egocentric view for the purpose of predicting intention and forecasting future trajectory. A challenge in addressing this problem in urban traffic scenes is attributed to the unpredictable behavior of pedestrians, whereby actions and intentions are constantly in flux and depend on the pedestrians pose, their 3D spatial relations, and their interaction with other agents as well as with the environment. To partially address these challenges, we consider the importance of pose toward recognition and 3D localization of pedestrian actions. In particular, we propose an action recognition framework using a two-stream temporal relation network with inputs corresponding to the raw RGB image sequence of the tracked pedestrian as well as the pedestrian pose. The proposed method outperforms methods using a single-stream temporal relation network based on evaluations using the JAAD public dataset. The estimated pose and associated body key-points are also used as input to a network that estimates the 3D location of the pedestrian using a unique loss function. The evaluation of our 3D localization method on the KITTI dataset indicates the improvement of the average localization error as compared to existing state-of-the-art methods. Finally, we conduct qualitative tests of action recognition and 3D localization on HRI's H3D driving dataset.
Deep Convolutional Neural Networks have proven effective in solving the task of semantic segmentation. However, their efficiency heavily relies on the pixel-level annotations that are expensive to get and often require domain expertise, especially in medical imaging. Weakly supervised semantic segmentation helps to overcome these issues and also provides explainable deep learning models. In this paper, we propose a novel approach to the semantic segmentation of medical chest X-ray images with only image-level class labels as supervision. We improve the disease localization accuracy by combining three approaches as consecutive steps. First, we generate pseudo segmentation labels of abnormal regions in the training images through a supervised classification model enhanced with a regularization procedure. The obtained activation maps are then post-processed and propagated into a second classification model-Inter-pixel Relation Network, which improves the boundaries between different object classes. Finally, the resulting pseudo-labels are used to train a proposed fully supervised segmentation model. We analyze the robustness of the presented method and test its performance on two distinct datasets: PASCAL VOC 2012 and SIIM-ACR Pneumothorax. We achieve significant results in the segmentation on both datasets using only image-level annotations. We show that this approach is applicable to chest X-rays for detecting an anomalous volume of air in the pleural space between the lung and the chest wall. Our code has been made publicly available.
Despite achieving state-of-the-art performance, deep learning methods generally require a large amount of labeled data during training and may suffer from overfitting when the sample size is small. To ensure good generalizability of deep networks under small sample sizes, learning discriminative features is crucial. To this end, several loss functions have been proposed to encourage large intra-class compactness and inter-class separability. In this paper, we propose to enhance the discriminative power of features from a new perspective by introducing a novel neural network termed Relation-and-Margin learning Network (ReMarNet). Our method assembles two networks of different backbones so as to learn the features that can perform excellently in both of the aforementioned two classification mechanisms. Specifically, a relation network is used to learn the features that can support classification based on the similarity between a sample and a class prototype; at the meantime, a fully connected network with the cross entropy loss is used for classification via the decision boundary. Experiments on four image datasets demonstrate that our approach is effective in learning discriminative features from a small set of labeled samples and achieves competitive performance against state-of-the-art methods. Codes are available at https://github.com/liyunyu08/ReMarNet.
This technical report introduces our winning solution to the spatio-temporal action localization track, AVA-Kinetics Crossover, in ActivityNet Challenge 2020. Our entry is mainly based on Actor-Context-Actor Relation Network. We describe technical details for the new AVA-Kinetics dataset, together with some experimental results. Without any bells and whistles, we achieved 39.62 mAP on the test set of AVA-Kinetics, which outperforms other entries by a large margin. Code will be available at: https://github.com/Siyu-C/ACAR-Net.
Localizing persons and recognizing their actions from videos is a challenging task towards high-level video understanding. Recent advances have been achieved by modeling direct pairwise relations between entities. In this paper, we take one step further, not only model direct relations between pairs but also take into account indirect higher-order relations established upon multiple elements. We propose to explicitly model the Actor-Context-Actor Relation, which is the relation between two actors based on their interactions with the context. To this end, we design an Actor-Context-Actor Relation Network (ACAR-Net) which builds upon a novel High-order Relation Reasoning Operator and an Actor-Context Feature Bank to enable indirect relation reasoning for spatio-temporal action localization. Experiments on AVA and UCF101-24 datasets show the advantages of modeling actor-context-actor relations, and visualization of attention maps further verifies that our model is capable of finding relevant higher-order relations to support action detection. Notably, our method ranks first in the AVA-Kineticsaction localization task of ActivityNet Challenge 2020, out-performing other entries by a significant margin (+6.71mAP). Training code and models will be available at https://github.com/Siyu-C/ACAR-Net.
Fashion is an inherently visual concept and computer vision and artificial intelligence (AI) are playing an increasingly important role in shaping the future of this domain. Many research has been done on recommending fashion products based on the learned user preferences. However, in addition to recommending single items, AI can also help users create stylish outfits from items they already have, or purchase additional items that go well with their current wardrobe. Compatibility is the key factor in creating stylish outfits from single items. Previous studies have mostly focused on modeling pair-wise compatibility. There are a few approaches that consider an entire outfit, but these approaches have limitations such as requiring rich semantic information, category labels, and fixed order of items. Thus, they fail to effectively determine compatibility when such information is not available. In this work, we adopt a Relation Network (RN) to develop new compatibility learning models, Fashion RN and FashionRN-VSE, that addresses the limitations of existing approaches. FashionRN learns the compatibility of an entire outfit, with an arbitrary number of items, in an arbitrary order. We evaluated our model using a large dataset of 49,740 outfits that we collected from Polyvore website. Quantitatively, our experimental results demonstrate state of the art performance compared with alternative methods in the literature in both compatibility prediction and fill-in-the-blank test. Qualitatively, we also show that the item embedding learned by FashionRN indicate the compatibility among fashion items.
Metric-based few-shot learning methods concentrate on learning transferable feature embedding that generalizes well from seen categories to unseen categories under the supervision of limited number of labelled instances. However, most of them treat each individual instance in the working context separately without considering its relationships with the others. In this work, we investigate a new metric-learning method, Memory-Augmented Relation Network (MRN), to explicitly exploit these relationships. In particular, for an instance, we choose the samples that are visually similar from the working context, and perform weighted information propagation to attentively aggregate helpful information from the chosen ones to enhance its representation. In MRN, we also formulate the distance metric as a learnable relation module which learns to compare for similarity measurement, and augment the working context with memory slots, both contributing to its generality. We empirically demonstrate that MRN yields significant improvement over its ancestor and achieves competitive or even better performance when compared with other few-shot learning approaches on the two major benchmark datasets, i.e. miniImagenet and tieredImagenet.
Brain-computer interface (BCI) is a communication tool that connects users and external devices. In a real-time BCI environment, a calibration procedure is particularly necessary for each user and each session. This procedure consumes a significant amount of time that hinders the application of a BCI system in a real-world scenario. To avoid this problem, we adopt the metric based few-shot learning approach for decoding intuitive upper-extremity movement imagination (MI) using a gradual relation network (GRN) that can gradually consider the combination of temporal and spectral groups. We acquired the MI data of the upper-arm, forearm, and hand associated with intuitive upper-extremity movement from 25 subjects. The grand average multiclass classification results under offline analysis were 42.57%, 55.60%, and 80.85% in 1-, 5-, and 25-shot settings, respectively. In addition, we could demonstrate the feasibility of intuitive MI decoding using the few-shot approach in real-time robotic arm control scenarios. Five participants could achieve a success rate of 78% in the drinking task. Hence, we demonstrated the feasibility of the online robotic arm control with shortened calibration time by focusing on human body parts but also the accommodation of various untrained intuitive MI decoding based on the proposed GRN.
Learning representations of spatial references in natural language is a key challenge in tasks like autonomous navigation and robotic manipulation. Recent work has investigated various neural architectures for learning multi-modal representations for spatial concepts. However, the lack of explicit reasoning over entities makes such approaches vulnerable to noise in input text or state observations. In this paper, we develop effective models for understanding spatial references in text that are robust and interpretable, without sacrificing performance. We design a text-conditioned \textit{relation network} whose parameters are dynamically computed with a cross-modal attention module to capture fine-grained spatial relations between entities. This design choice provides interpretability of learned intermediate outputs. Experiments across three tasks demonstrate that our model achieves superior performance, with a 17\% improvement in predicting goal locations and a 15\% improvement in robustness compared to state-of-the-art systems.
Existing work on augmenting question answering (QA) models with external knowledge (e.g., knowledge graphs) either struggle to model multi-hop relations efficiently, or lack transparency into the model's prediction rationale. In this paper, we propose a novel knowledge-aware approach that equips pre-trained language models (PTLMs) with a multi-hop relational reasoning module, named multi-hop graph relation network (MHGRN). It performs multi-hop, multi-relational reasoning over subgraphs extracted from external knowledge graphs. The proposed reasoning module unifies path-based reasoning methods and graph neural networks to achieve better interpretability and scalability. We also empirically show its effectiveness and scalability on CommonsenseQA and OpenbookQA datasets, and interpret its behaviors with case studies.
Multi-task learning (mtl) provides state-of-the-art results in many applications of computer vision and natural language processing. In contrast to single-task learning (stl), mtl allows for leveraging knowledge between related tasks improving prediction results on the main task (in contrast to an auxiliary task) or all tasks. However, there is a limited number of comparative studies on applying mtl architectures for regression and time series problems taking recent advances of mtl into account. An interesting, non-linear problem is the forecast of the expected power generation for renewable power plants. Therefore, this article provides a comparative study of the following recent and important mtl architectures: Hard parameter sharing, cross-stitch network, sluice network (sn). They are compared to a multi-layer perceptron model of similar size in an stl setting. Additionally, we provide a simple, yet effective approach to model task specific information through an embedding layer in an multi-layer perceptron, referred to as task embedding. Further, we introduce a new mtl architecture named emerging relation network (ern), which can be considered as an extension of the sluice network. For a solar power dataset, the task embedding achieves the best mean improvement with 14.9%. The mean improvement of the ern and the sn on the solar dataset is of similar magnitude with 14.7% and 14.8%. On a wind power dataset, only the ern achieves a significant improvement of up to 7.7%. Results suggest that the ern is beneficial when tasks are only loosely related and the prediction problem is more non-linear. Contrary, the proposed task embedding is advantageous when tasks are strongly correlated. Further, the task embedding provides an effective approach with reduced computational effort compared to other mtl architectures.
The training of deep-learning-based text classification models relies heavily on a huge amount of annotation data, which is difficult to obtain. When the labeled data is scarce, models tend to struggle to achieve satisfactory performance. However, human beings can distinguish new categories very efficiently with few examples. This is mainly due to the fact that human beings can leverage knowledge obtained from relevant tasks. Inspired by human intelligence, we propose to introduce external knowledge into few-shot learning to imitate human knowledge. A novel parameter generator network is investigated to this end, which is able to use the external knowledge to generate relation network parameters. Metrics can be transferred among tasks when equipped with these generated parameters, so that similar tasks use similar metrics while different tasks use different metrics. Through experiments, we demonstrate that our method outperforms the state-of-the-art few-shot text classification models.
Video question answering (VideoQA) is challenging as it requires modeling capacity to distill dynamic visual artifacts and distant relations and to associate them with linguistic concepts. We introduce a general-purpose reusable neural unit called Conditional Relation Network (CRN) that serves as a building block to construct more sophisticated structures for representation and reasoning over video. CRN takes as input an array of tensorial objects and a conditioning feature, and computes an array of encoded output objects. Model building becomes a simple exercise of replication, rearrangement and stacking of these reusable units for diverse modalities and contextual information. This design thus supports high-order relational and multi-step reasoning. The resulting architecture for VideoQA is a CRN hierarchy whose branches represent sub-videos or clips, all sharing the same question as the contextual condition. Our evaluations on well-known datasets achieved new SoTA results, demonstrating the impact of building a general-purpose reasoning unit on complex domains such as VideoQA.
Dense indoor scene modeling from 2D images has been bottlenecked due to the absence of depth information and cluttered occlusions. We present an automatic indoor scene modeling approach using deep features from neural networks. Given a single RGB image, our method simultaneously recovers semantic contents, 3D geometry and object relationship by reasoning indoor environment context. Particularly, we design a shallow-to-deep architecture on the basis of convolutional networks for semantic scene understanding and modeling. It involves multi-level convolutional networks to parse indoor semantics/geometry into non-relational and relational knowledge. Non-relational knowledge extracted from shallow-end networks (e.g. room layout, object geometry) is fed forward into deeper levels to parse relational semantics (e.g. support relationship). A Relation Network is proposed to infer the support relationship between objects. All the structured semantics and geometry above are assembled to guide a global optimization for 3D scene modeling. Qualitative and quantitative analysis demonstrates the feasibility of our method in understanding and modeling semantics-enriched indoor scenes by evaluating the performance of reconstruction accuracy, computation performance and scene complexity.
Though recent research has achieved remarkable progress in generating realistic images with generative adversarial networks (GANs), the lack of training stability is still a lingering concern of most GANs, especially on high-resolution inputs and complex datasets. Since the randomly generated distribution can hardly overlap with the real distribution, training GANs often suffers from the gradient vanishing problem. A number of approaches have been proposed to address this issue by constraining the discriminator's capabilities using empirical techniques, like weight clipping, gradient penalty, spectral normalization etc. In this paper, we provide a more principled approach as an alternative solution to this issue. Instead of training the discriminator to distinguish real and fake input samples, we investigate the relationship between paired samples by training the discriminator to separate paired samples from the same distribution and those from different distributions. To this end, we explore a relation network architecture for the discriminator and design a triplet loss which performs better generalization and stability. Extensive experiments on benchmark datasets show that the proposed relation discriminator and new loss can provide significant improvement on variable vision tasks including unconditional and conditional image generation and image translation.
Cyber-physical system (CPS) has operated, controlled and coordinated the physical systems integrated by a computing and communication core applied in industry 4.0. To accommodate CPS services, fog radio and optical networks (F-RON) has become an important supporting physical cyber infrastructure taking advantage of both the inherent ubiquity of wireless technology and the large capacity of optical networks. However, cyber security is the biggest issue in CPS scenario as there is a tradeoff between security control and privacy exposure in F-RON. To deal with this issue, we propose a brain-like based distributed control security (BLCS) architecture for F-RON in CPS, by introducing a brain-like security (BLS) scheme. BLCS can accomplish the secure cross-domain control among tripartite controllers verification in the scenario of decentralized F-RON for distributed computing and communications, which has no need to disclose the private information of each domain against cyber-attacks. BLS utilizes parts of information to perform control identification through relation network and deep learning of behavior library. The functional modules of BLCS architecture are illustrated including various controllers and brain-like knowledge base. The interworking procedures in distributed control security modes based on BLS are described. The overall feasibility and efficiency of architecture are experimentally verified on the software defined network testbed in terms of average mistrust rate, path provisioning latency, packet loss probability and blocking probability. The emulation results are obtained and dissected based on the testbed.
Person re-identification (reID) aims at retrieving an image of the person of interest from a set of images typically captured by multiple cameras. Recent reID methods have shown that exploiting local features describing body parts, together with a global feature of a person image itself, gives robust feature representations, even in the case of missing body parts. However, using the individual part-level features directly, without considering relations between body parts, confuses differentiating identities of different persons having similar attributes in corresponding parts. To address this issue, we propose a new relation network for person reID that considers relations between individual body parts and the rest of them. Our model makes a single part-level feature incorporate partial information of other body parts as well, supporting it to be more discriminative. We also introduce a global contrastive pooling (GCP) method to obtain a global feature of a person image. We propose to use contrastive features for GCP to complement conventional max and averaging pooling techniques. We show that our model outperforms the state of the art on the Market1501, DukeMTMC-reID and CUHK03 datasets, demonstrating the effectiveness of our approach on discriminative person representations.
Attention mechanisms have been boosting the performance of deep learning models on a wide range of applications, ranging from speech understanding to program induction. However, despite experiments from psychology which suggest that attention plays an essential role in visual reasoning, the full potential of attention mechanisms has so far not been explored to solve abstract cognitive tasks on image data. In this work, we propose a hybrid network architecture, grounded on self-attention and relational reasoning. We call this new model Attention Relation Network (ARNe). ARNe combines features from the recently introduced Transformer and the Wild Relation Network (WReN). We test ARNe on the Procedurally Generated Matrices (PGMs) datasets for abstract visual reasoning. ARNe excels the WReN model on this task by 11.28 ppt. Relational concepts between objects are efficiently learned demanding only 35% of the training samples to surpass reported accuracy of the base line model. Our proposed hybrid model, represents an alternative on learning abstract relations using self-attention and demonstrates that the Transformer network is also well suited for abstract visual reasoning.