Semantic image understanding is a challenging topic in computer vision. It requires to detect all objects in an image, but also to identify all the relations between them. Detected objects, their labels and the discovered relations can be used to construct a scene graph which provides an abstract semantic interpretation of an image. In previous works, relations were identified by solving an assignment problem formulated as Mixed-Integer Linear Programs. In this work, we interpret that formulation as Ordinary Differential Equation (ODE). The proposed architecture performs scene graph inference by solving a neural variant of an ODE by end-to-end learning. It achieves state-of-the-art results on all three benchmark tasks: scene graph generation (SGGen), classification (SGCls) and visual relationship detection (PredCls) on Visual Genome benchmark.
We address the problem of Visual Relationship Detection (VRD) which aims to describe the relationships between pairs of objects in the form of triplets of (subject, predicate, object). We observe that given a pair of bounding box proposals, objects often participate in multiple relations implying the distribution of triplets is multimodal. We leverage the strong correlations within triplets to learn the joint distribution of triplet variables conditioned on the image and the bounding box proposals, doing away with the hitherto used independent distribution of triplets. To make learning the triplet joint distribution feasible, we introduce a novel technique of learning conditional triplet distributions in the form of their normalized low rank non-negative tensor decompositions. Normalized tensor decompositions take form of mixture distributions of discrete variables and thus are able to capture multimodality. This allows us to efficiently learn higher order discrete multimodal distributions and at the same time keep the parameter size manageable. We further model the probability of selecting an object proposal pair and include a relation triplet prior in our model. We show that each part of the model improves performance and the combination outperforms state-of-the-art score on the Visual Genome (VG) and Visual Relationship Detection (VRD) datasets.
Visual relationship detection, as a challenging task used to find and distinguish the interactions between object pairs in one image, has received much attention recently. In this work, we propose a novel visual relationship detection framework by deeply mining and utilizing relative location of object-pair in every stage of the procedure. In both the stages, relative location information of each object-pair is abstracted and encoded as auxiliary feature to improve the distinguishing capability of object-pairs proposing and predicate recognition, respectively; Moreover, one Gated Graph Neural Network(GGNN) is introduced to mine and measure the relevance of predicates using relative location. With the location-based GGNN, those non-exclusive predicates with similar spatial position can be clustered firstly and then be smoothed with close classification scores, thus the accuracy of top $n$ recall can be increased further. Experiments on two widely used datasets VRD and VG show that, with the deeply mining and exploiting of relative location information, our proposed model significantly outperforms the current state-of-the-art.
One of the most difficult tasks in scene understanding is recognizing interactions between objects in an image. This task is often called visual relationship detection (VRD). We consider the question of whether, given auxiliary textual data in addition to the standard visual data used for training VRD models, VRD performance can be improved. We present a new deep model that can leverage additional textual data. Our model relies on a shared text--image representation of subject-verb-object relationships appearing in the text, and object interactions in images. Our method is the first to enable recognition of visual relationships missing in the visual training data and appearing only in the auxiliary text. We test our approach on two different text sources: text originating in images and text originating in books. We test and validate our approach using two large-scale recognition tasks: VRD and Scene Graph Generation. We show a surprising result: Our approach works better with text originating in books, and outperforms the text originating in images on the task of unseen relationship recognition. It is comparable to the model which utilizes text originating in images on the task of seen relationship recognition.
Human-object interaction detection is an important and relatively new class of visual relationship detection tasks, essential for deeper scene understanding. Most existing approaches decompose the problem into object localization and interaction recognition. Despite showing progress, these approaches only rely on the appearances of humans and objects and overlook the available context information, crucial for capturing subtle interactions between them. We propose a contextual attention framework for human-object interaction detection. Our approach leverages context by learning contextually-aware appearance features for human and object instances. The proposed attention module then adaptively selects relevant instance-centric context information to highlight image regions likely to contain human-object interactions. Experiments are performed on three benchmarks: V-COCO, HICO-DET and HCVRD. Our approach outperforms the state-of-the-art on all datasets. On the V-COCO dataset, our method achieves a relative gain of 4.4% in terms of role mean average precision ($mAP_{role}$), compared to the existing best approach.
Semantic Image Interpretation is the task of extracting a structured semantic description from images. This requires the detection of visual relationships: triples (subject,relation,object) describing a semantic relation between a subject and an object. A pure supervised approach to visual relationship detection requires a complete and balanced training set for all the possible combinations of (subject, relation, object). However, such training sets are not available and would require a prohibitive human effort. This implies the ability of predicting triples which do not appear in the training set. This problem is called zero-shot learning. State-of-the-art approaches to zero-shot learning exploit similarities among relationships in the training set or external linguistic knowledge. In this paper, we perform zero-shot learning by using Logic Tensor Networks, a novel Statistical Relational Learning framework that exploits both the similarities with other seen relationships and background knowledge, expressed with logical constraints between subjects, relations and objects. The experiments on the Visual Relationship Dataset show that the use of logical constraints outperforms the current methods. This implies that background knowledge can be used to alleviate the incompleteness of training sets.
Relations amongst entities play a central role in image understanding. Due to the complexity of modeling (subject, predicate, object) relation triplets, it is crucial to develop a method that can not only recognize seen relations, but also generalize to unseen cases. Inspired by a previously proposed visual translation embedding model, or VTransE, we propose a context-augmented translation embedding model that can capture both common and rare relations. The previous VTransE model maps entities and predicates into a low-dimensional embedding vector space where the predicate is interpreted as a translation vector between the embedded features of the bounding box regions of the subject and the object. Our model additionally incorporates the contextual information captured by the bounding box of the union of the subject and the object, and learns the embeddings guided by the constraint predicate $\approx$ union (subject, object) $-$ subject $-$ object. In a comprehensive evaluation on multiple challenging benchmarks, our approach outperforms previous translation-based models and comes close to or exceeds the state of the art across a range of settings, from small-scale to large-scale datasets, from common to previously unseen relations. It also achieves promising results for the recently introduced task of scene graph generation.
In visual relationship detection, human-notated relationships can be regarded as determinate relationships. However, there are still large amount of unlabeled data, such as object pairs with less significant relationships or even with no relationships. We refer to these unlabeled but potentially useful data as undetermined relationships. Although a vast body of literature exists, few methods exploit these undetermined relationships for visual relationship detection. In this paper, we explore the beneficial effect of undetermined relationships on visual relationship detection. We propose a novel multi-modal feature based undetermined relationship learning network (MF-URLN) and achieve great improvements in relationship detection. In detail, our MF-URLN automatically generates undetermined relationships by comparing object pairs with human-notated data according to a designed criterion. Then, the MF-URLN extracts and fuses features of object pairs from three complementary modals: visual, spatial, and linguistic modals. Further, the MF-URLN proposes two correlated subnetworks: one subnetwork decides the determinate confidence, and the other predicts the relationships. We evaluate the MF-URLN on two datasets: the Visual Relationship Detection (VRD) and the Visual Genome (VG) datasets. The experimental results compared with state-of-the-art methods verify the significant improvements made by the undetermined relationships, e.g., the top-50 relation detection recall improves from 19.5% to 23.9% on the VRD dataset.
Visual relation detection methods rely on object information extracted from RGB images such as 2D bounding boxes, feature maps, and predicted class probabilities. We argue that depth maps can additionally provide valuable information on object relations, e.g. helping to detect not only spatial relations, such as standing behind, but also non-spatial relations, such as holding. In this work, we study the effect of using different object features with a focus on depth maps. To enable this study, we release a new synthetic dataset of depth maps, VG-Depth, as an extension to Visual Genome (VG). We also note that given the highly imbalanced distribution of relations in VG, typical evaluation metrics for visual relation detection cannot reveal improvements of under-represented relations. To address this problem, we propose using an additional metric, calling it Macro Recall@K, and demonstrate its remarkable performance on VG. Finally, our experiments confirm that by effective utilization of depth maps within a simple, yet competitive framework, the performance of visual relation detection can be improved by a margin of up to 8%.
Visual relationship detection is an intermediate image understanding task that detects two objects and classifies a predicate that explains the relationship between two objects in an image. The three components are linguistically and visually correlated (e.g. "wear" is related to "person" and "shirt", while "laptop" is related to "table" and "on") thus, the solution space is huge because there are many possible cases between them. Language and visual modules are exploited and a sophisticated spatial vector is proposed. The models in this work outperformed the state of arts without costly linguistic knowledge distillation from a large text corpus and building complex loss functions. All experiments were only evaluated on Visual Relationship Detection and Visual Genome dataset.
Scene graph generation aims to provide a semantic and structural description of an image, denoting the objects (with nodes) and their relationships (with edges). The best performing works to date are based on exploiting the context surrounding objects or relations,e.g., by passing information among objects. In these approaches, to transform the representation of source objects is a critical process for extracting information for the use by target objects. In this work, we argue that a source object should give what tar-get object needs and give different objects different information rather than contributing common information to all targets. To achieve this goal, we propose a Target-TailoredSource-Transformation (TTST) method to efficiently propagate information among object proposals and relations. Particularly, for a source object proposal which will contribute information to other target objects, we transform the source object feature to the target object feature domain by simultaneously taking both the source and target into account. We further explore more powerful representations by integrating language prior with the visual context in the transformation for the scene graph generation. By doing so the target object is able to extract target-specific information from the source object and source relation accordingly to refine its representation. Our framework is validated on the Visual Genome bench-mark and demonstrated its state-of-the-art performance for the scene graph generation. The experimental results show that the performance of object detection and visual relation-ship detection are promoted mutually by our method.
Scene graph generation has received growing attention with the advancements in image understanding tasks such as object detection, attributes and relationship prediction,~\etc. However, existing datasets are biased in terms of object and relationship labels, or often come with noisy and missing annotations, which makes the development of a reliable scene graph prediction model very challenging. In this paper, we propose a novel scene graph generation algorithm with external knowledge and image reconstruction loss to overcome these dataset issues. In particular, we extract commonsense knowledge from the external knowledge base to refine object and phrase features for improving generalizability in scene graph generation. To address the bias of noisy object annotations, we introduce an auxiliary image reconstruction path to regularize the scene graph generation network. Extensive experiments show that our framework can generate better scene graphs, achieving the state-of-the-art performance on two benchmark datasets: Visual Relationship Detection and Visual Genome datasets.
Visual Relationship Detection is defined as, given an image composed of a subject and an object, the correct relation is predicted. To improve the visual part of this difficult problem, ten preprocessing methods were tested to determine whether the widely used Union method yields the optimal results. Therefore, focusing solely on predicate prediction, no object detection and linguistic knowledge were used to prevent them from affecting the comparison results. Once fine-tuned, the Visual Geometry Group models were evaluated using Recall@1, per-predicate recall, activation maximisations, class activation maps, and error analysis. From this research it was found that using preprocessing methods such as the Union-Without-Background-and-with-Binary-mask (Union-WB-and-B) method yields significantly better results than the widely used Union method since, as designed, it enables the Convolutional Neural Network to also identify the subject and object in the convolutional layers instead of solely in the fully-connected layers.
In this paper we investigate the problems of class imbalance and irrelevant relationships in Visual Relationship Detection (VRD). State-of-the-art deep VRD models still struggle to predict uncommon classes, limiting their applicability. Moreover, many methods are incapable of properly filtering out background relationships while predicting relevant ones. Although these problems are very apparent, they have both been overlooked so far. We analyse why this is the case and propose modifications to both model and training to alleviate the aforementioned issues, as well as suggesting new measures to complement existing ones and give a more holistic picture of the efficacy of a model.
Visual semantic information comprises two important parts: the meaning of each visual semantic unit and the coherent visual semantic relation conveyed by these visual semantic units. Essentially, the former one is a visual perception task while the latter one corresponds to visual context reasoning. Remarkable advances in visual perception have been achieved due to the success of deep learning. In contrast, visual semantic information pursuit, a visual scene semantic interpretation task combining visual perception and visual context reasoning, is still in its early stage. It is the core task of many different computer vision applications, such as object detection, visual semantic segmentation, visual relationship detection or scene graph generation. Since it helps to enhance the accuracy and the consistency of the resulting interpretation, visual context reasoning is often incorporated with visual perception in current deep end-to-end visual semantic information pursuit methods. However, a comprehensive review for this exciting area is still lacking. In this survey, we present a unified theoretical paradigm for all these methods, followed by an overview of the major developments and the future trends in each potential direction. The common benchmark datasets, the evaluation metrics and the comparisons of the corresponding methods are also introduced.
Most scene graph parsers use a two-stage pipeline to detect visual relationships: the first stage detects entities, and the second predicts the predicate for each entity pair using a softmax distribution. We find that such pipelines, trained with only a cross entropy loss over predicate classes, suffer from two common errors. The first, Entity Instance Confusion, occurs when the model confuses multiple instances of the same type of entity (e.g. multiple cups). The second, Proximal Relationship Ambiguity, arises when multiple subject-predicate-object triplets appear in close proximity with the same predicate, and the model struggles to infer the correct subject-object pairings (e.g. mis-pairing musicians and their instruments). We propose a set of contrastive loss formulations that specifically target these types of errors within the scene graph parsing problem, collectively termed the Graphical Contrastive Losses. These losses explicitly force the model to disambiguate related and unrelated instances through margin constraints specific to each type of confusion. We further construct a relationship detector, called RelDN, using the aforementioned pipeline to demonstrate the efficacy of our proposed losses. Our model outperforms the winning method of the OpenImages Relationship Detection Challenge by 4.7\% (16.5\% relative) on the test set. We also show improved results over the best previous methods on the Visual Genome and Visual Relationship Detection datasets.
Detecting visual relationships, i.e. <Subject, Predicate, Object> triplets, is a challenging Scene Understanding task approached in the past via linguistic priors or spatial information in a single feature branch. We introduce a new deeply supervised two-branch architecture, the Multimodal Attentional Translation Embeddings, where the visual features of each branch are driven by a multimodal attentional mechanism that exploits spatio-linguistic similarities in a low-dimensional space. We present a variety of experiments comparing against all related approaches in the literature, as well as by re-implementing and fine-tuning several of them. Results on the commonly employed VRD dataset [1] show that the proposed method clearly outperforms all others, while we also justify our claims both quantitatively and qualitatively.
Multimodal representation learning is gaining more and more interest within the deep learning community. While bilinear models provide an interesting framework to find subtle combination of modalities, their number of parameters grows quadratically with the input dimensions, making their practical implementation within classical deep learning pipelines challenging. In this paper, we introduce BLOCK, a new multimodal fusion based on the block-superdiagonal tensor decomposition. It leverages the notion of block-term ranks, which generalizes both concepts of rank and mode ranks for tensors, already used for multimodal fusion. It allows to define new ways for optimizing the tradeoff between the expressiveness and complexity of the fusion model, and is able to represent very fine interactions between modalities while maintaining powerful mono-modal representations. We demonstrate the practical interest of our fusion model by using BLOCK for two challenging tasks: Visual Question Answering (VQA) and Visual Relationship Detection (VRD), where we design end-to-end learnable architectures for representing relevant interactions between modalities. Through extensive experiments, we show that BLOCK compares favorably with respect to state-of-the-art multimodal fusion models for both VQA and VRD tasks. Our code is available at https://github.com/Cadene/block.bootstrap.pytorch.
Visual relationship detection can bridge the gap between computer vision and natural language for scene understanding of images. Different from pure object recognition tasks, the relation triplets of subject-predicate-object lie on an extreme diversity space, such as \textit{person-behind-person} and \textit{car-behind-building}, while suffering from the problem of combinatorial explosion. In this paper, we propose a context-dependent diffusion network (CDDN) framework to deal with visual relationship detection. To capture the interactions of different object instances, two types of graphs, word semantic graph and visual scene graph, are constructed to encode global context interdependency. The semantic graph is built through language priors to model semantic correlations across objects, whilst the visual scene graph defines the connections of scene objects so as to utilize the surrounding scene information. For the graph-structured data, we design a diffusion network to adaptively aggregate information from contexts, which can effectively learn latent representations of visual relationships and well cater to visual relationship detection in view of its isomorphic invariance to graphs. Experiments on two widely-used datasets demonstrate that our proposed method is more effective and achieves the state-of-the-art performance.
We propose an efficient and interpretable scene graph generator. We consider three types of features: visual, spatial and semantic, and we use a late fusion strategy such that each feature's contribution can be explicitly investigated. We study the key factors about these features that have the most impact on the performance, and also visualize the learned visual features for relationships and investigate the efficacy of our model. We won the champion of the OpenImages Visual Relationship Detection Challenge on Kaggle, where we outperform the 2nd place by 5\% (20\% relatively). We believe an accurate scene graph generator is a fundamental stepping stone for higher-level vision-language tasks such as image captioning and visual QA, since it provides a semantic, structured comprehension of an image that is beyond pixels and objects.