Conventional phrase grounding aims to localize noun phrases mentioned in a given caption to their corresponding image regions, which has achieved great success recently. Apparently, sole noun phrase grounding is not enough for cross-modal visual language understanding. Here we extend the task by considering pronouns as well. First, we construct a dataset of phrase grounding with both noun phrases and pronouns to image regions. Based on the dataset, we test the performance of phrase grounding by using a state-of-the-art literature model of this line. Then, we enhance the baseline grounding model with coreference information which should help our task potentially, modeling the coreference structures with graph convolutional networks. Experiments on our dataset, interestingly, show that pronouns are easier to ground than noun phrases, where the possible reason might be that these pronouns are much less ambiguous. Additionally, our final model with coreference information can significantly boost the grounding performance of both noun phrases and pronouns.
University of Southern California（南加州大学）；Facebook Inc.（脸书公司）
A phrase grounding system localizes a particular object in an image referred to by a natural language query. In previous work, the phrases were restricted to have nouns that were encountered in training, we extend the task to Zero-Shot Grounding(ZSG) which can include novel, "unseen" nouns. Current phrase grounding systems use an explicit object detection network in a 2-stage framework where one stage generates sparse proposals and the other stage evaluates them. In the ZSG setting, generating appropriate proposals itself becomes an obstacle as the proposal generator is trained on the entities…
Given a natural language query, a phrase grounding system aims to localize mentioned objects in an image. In weakly supervised scenario, mapping between image regions (i.e., proposals) and language is not available in the training set. Previous methods address this deficiency by training a grounding system via learning to reconstruct language information contained in input queries from predicted proposals. However, the optimization is solely guided by the reconstruction loss from the language modality, and ignores rich visual information contained in proposals and useful cues from external kno…
University of Southern California（南加州大学）；Institute for Robotics and Intelligent Systems（机器人与智能系统研究所）
Given a textual description of an image, phrase grounding localizes objects in the image referred by query phrases in the description. State-of-the-art methods address the problem by ranking a set of proposals based on the relevance to each query, which are limited by the performance of independent proposal generation systems and ignore useful cues from context in the description. In this paper, we adopt a spatial regression method to break the performance limit, and introduce reinforcement learning techniques to further leverage semantic context information. We propose a novel Query-guided Re…
In this paper, we propose a novel graph learning framework for phrase grounding in the image. Developing from the sequential to the dense graph model, existing works capture coarse-grained context but fail to distinguish the diversity of context among phrases and image regions. In contrast, we pay special attention to different motifs implied in the context of the scene graph and devise the disentangled graph network to integrate the motif-aware contextual information into representations. Besides, we adopt interventional strategies at the feature and the structure levels to consolidate and ge…
Max Planck Institute for Informatics（马克斯·普朗克信息学研究所）；UC Berkeley（加州大学伯克利分校）；ICSI（国际计算机科学研究所）
Grounding (i.e. localizing) arbitrary, free-form textual phrases in visual content is a challenging problem with many applications for human-computer interaction and image-text reference resolution. Few datasets provide the ground truth spatial localization of phrases, thus it is desirable to learn from data with no or little grounding supervision. We propose a novel approach which learns grounding by reconstructing a given phrase using an attention mechanism, which can be either latent or optimized directly. During training our approach encodes the phrase using a recurrent network language mo…
We present a new problem: grounding natural language instructions to mobile user interface actions, and create three new datasets for it. For full task evaluation, we create PIXELHELP, a corpus that pairs English instructions with actions performed by people on a mobile UI emulator. To scale training, we decouple the language and action data by (a) annotating action phrase spans in HowTo instructions and (b) synthesizing grounded descriptions of actions for mobile user interfaces. We use a Transformer to extract action phrase tuples from long-range natural language instructions. A grounding Tr…
Grounding free-form textual queries necessitates an understanding of these textual phrases and its relation to the visual cues to reliably reason about the described locations. Spatial attention networks are known to learn this relationship and focus its gaze on salient objects in the image. Thus, we propose to utilize spatial attention networks for image-level visual-textual fusion preserving local (word) and global (phrase) information to refine region proposals with an in-network Region Proposal Network (RPN) and detect single or multiple regions for a phrase query. We focus only on the phr…
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；NVIDIA（英伟达）；Bar Ilan University（巴伊兰大学）
Phrase grounding, the problem of associating image regions to caption words, is a crucial component of vision-language tasks. We show that phrase grounding can be learned by optimizing word-region attention to maximize a lower bound on mutual information between images and caption words. Given pairs of images and captions, we maximize compatibility of the attention-weighted regions and the words in the corresponding caption, compared to non-corresponding pairs of images and captions. A key idea is to construct effective negative captions for learning through language model guided word substitu…
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
The phrase grounding task aims to ground each entity mention in a given caption of an image to a corresponding region in that image. Although there are clear dependencies between how different mentions of the same caption should be grounded, previous structured prediction methods that aim to capture such dependencies need to resort to approximate inference or non-differentiable losses. In this paper, we formulate phrase grounding as a sequence labeling task where we treat candidate regions as potential labels, and use neural chain Conditional Random Fields (CRFs) to model dependencies among re…
Boosting Temporal Sentence Grounding via Causal Inference
School of Electronic Engineering, Xidian University Xi'an China；School of Information Science \& Engineering, Lanzhou University Lanzhou China；Xidian University；Lanzhou University
Shanghai Jiao Tong University（上海交通大学）；UNC Chapel Hill（北卡罗来纳大学教堂山分校）；University of California, Berkeley（加州大学伯克利分校）
Phrase localization is a task that studies the mapping from textual phrases to regions of an image. Given difficulties in annotating phrase-to-object datasets at scale, we develop a Multimodal Alignment Framework (MAF) to leverage more widely-available caption-image datasets, which can then be used as a form of weak supervision. We first present algorithms to model phrase-object relevance by leveraging fine-grained visual representations and visually-aware language representations. By adopting a contrastive objective, our method uses information in caption-image pairs to boost the performance …
In this work, we propose Masked Noun-Phrase Prediction (MNPP), a pre-training strategy to tackle pronoun resolution in a fully unsupervised setting. Firstly, We evaluate our pre-trained model on various pronoun resolution datasets without any finetuning. Our method outperforms all previous unsupervised methods on all datasets by large margins. Secondly, we proceed to a few-shot setting where we finetune our pre-trained model on WinoGrande-S and XS separately. Our method outperforms RoBERTa-large baseline with large margins, meanwhile, achieving a higher AUC score after further finetuning on th…
PANORAMA: Panoptic Grounded Captioning via Mask Proposal Selection
Inria, École normale supérieure, CNRS, PSL Research University（法国国家信息与自动化研究所，巴黎高等师范学院，法国国家科学研究中心，巴黎文理研究大学）；Czech Institute of Informatics, Robotics and Cybernetics, Czech Technical University in Prague（捷克信息学、机器人与控制论研究所，布拉格捷克理工大学）
School of Informatics, University of Edinburgh（爱丁堡大学信息学院）；CFAR, IHPC, A*STAR（新加坡科技研究局高性能计算研究所认知智能研究中心）
Coreference resolution aims to identify words and phrases which refer to same entity in a text, a core task in natural language processing. In this paper, we extend this task to resolving coreferences in long-form narrations of visual scenes. First we introduce a new dataset with annotated coreference chains and their bounding boxes, as most existing image-text datasets only contain short sentences without coreferring expressions or labeled chains. We propose a new technique that learns to identify coreference chains using weak supervision, only from image-text pairs and a regularization using…
Tsinghua Shenzhen International Graduate School（清华大学深圳国际研究生院）；Meituan（美团）；Tsinghua University（清华大学）；Columbia University（哥伦比亚大学）
Temporal Sentence Grounding in Videos (TSGV), which aims to ground a natural language sentence in an untrimmed video, has drawn widespread attention over the past few years. However, recent studies have found that current benchmark datasets may have obvious moment annotation biases, enabling several simple baselines even without training to achieve SOTA performance. In this paper, we take a closer look at existing evaluation protocols, and find both the prevailing dataset and evaluation metrics are the devils that lead to untrustworthy benchmarking. Therefore, we propose to re-organize the two…
We address the problem of phrase grounding by lear ing a multi-level common semantic space shared by the textual and visual modalities. We exploit multiple levels of feature maps of a Deep Convolutional Neural Network, as well as contextualized word and sentence embeddings extracted from a character-based language model. Following dedicated non-linear mappings for visual features at each level, word, and sentence embeddings, we obtain multiple instantiations of our common semantic space in which comparisons between any target text and the visual content is performed with cosine similarity. We …
Possessive pronouns are used as determiners in English when no equivalent would be used in a Japanese sentence with the same meaning. This paper proposes a heuristic method of generating such possessive pronouns even when there is no equivalent in the Japanese. The method uses information about the use of possessive pronouns in English treated as a lexical property of nouns, in addition to contextual information about noun phrase referentiality and the subject and main verb of the sentence that the noun phrase appears in. The proposed method has been implemented in NTT Communication Science La…
Hierarchical Local-Global Transformer for Temporal Sentence Grounding
Hubei Engineering Research Center on Big Data Security, School of Cyber Science and Engineering, Huazhong University of Science and Technology（大数据安全湖北工程研究中心，华中科技大学网络安全科学与工程学院）；Wangxuan Institute of Computer fTechnology, Peking University（王宣计算机技术研究院，北京大学）；School of software, Dalian University of Technology（软件学院，大连理工大学）；School of Computer Science, and Technology, Huazhong University of Science, and Technology（计算机科学与技术学院，华中科技大学）
While real world challenges typically define visual categories with language words or phrases, most visual classification methods define categories with numerical indices. However, the language specification of the classes provides an especially useful prior for biased and noisy datasets, where it can help disambiguate what features are task-relevant. Recently, large-scale multimodal models have been shown to recognize a wide variety of high-level concepts from a language specification even without additional image training data, but they are often unable to distinguish classes for more fine-g…
Robots are widely collaborating with human users in diferent tasks that require high-level cognitive functions to make them able to discover the surrounding environment. A difcult challenge that we briefy highlight in this short paper is inferring the latent grammatical structure of language, which includes grounding parts of speech (e.g., verbs, nouns, adjectives, and prepositions) through visual perception, and induction of Combinatory Categorial Grammar (CCG) for phrases. This paves the way towards grounding phrases so as to make a robot able to understand human instructions appropriately d…
University of Edinburgh（爱丁堡大学）；University of Amsterdam（阿姆斯特丹大学）
Exploiting visual groundings for language understanding has recently been drawing much attention. In this work, we study visually grounded grammar induction and learn a constituency parser from both unlabeled text and its visual groundings. Existing work on this task (Shi et al., 2019) optimizes a parser via Reinforce and derives the learning signal only from the alignment of images and sentences. While their model is relatively accurate overall, its error distribution is very uneven, with low performance on certain constituents types (e.g., 26.2% recall on verb phrases, VPs) and high on other…
International Institute of Information Technology, Hyderabad（海得拉巴国际信息技术学院）
In this paper, we present a simple baseline for visual grounding for autonomous driving which outperforms the state of the art methods, while retaining minimal design choices. Our framework minimizes the cross-entropy loss over the cosine distance between multiple image ROI features with a text embedding (representing the give sentence/phrase). We use pre-trained networks for obtaining the initial embeddings and learn a transformation layer on top of the text embedding. We perform experiments on the Talk2Car dataset and achieve 68.7% AP50 accuracy, improving upon the previous state of the art …
Czech Institute of Informatics, Robotics and Cybernetics at the Czech Technical University in Prague（布拉格捷克理工大学捷克信息学、机器人学与控制论研究所）；Inria, École normale supérieure, CNRS, PSL Research University（法国国家信息与自动化研究所、巴黎高等师范学院、法国国家科学研究中心、PSL大学）
本文提出了grounded video caption generation新任务，构建了GROC测试集和HowToGround训练集，并开发了VideoGround模型，通过统一视频字幕生成与目标定位，在时间一致的边界框基础上实现了该任务的最新技术水平。
We develop a system to disambiguate object instances within the same class based on simple physical descriptions. The system takes as input a natural language phrase and a depth image containing a segmented object and predicts how similar the observed object is to the object described by the phrase. Our system is designed to learn from only a small amount of human-labeled language data and generalize to viewpoints not represented in the language-annotated depth image training set. By decoupling 3D shape representation from language representation, this method is able to ground language to nove…
Disambiguating Reference in Visually Grounded Dialogues through Joint Modeling of Textual and Multimodal Semantic Structures
Nara Institute of Science and Technology（奈良科学技術研究所）；Guardian Robot Project（守護機器人計劃）；RIKEN（理化学研究所）；Kyoto University（京都大學）；Institute of Science Tokyo（東京科學研究院）
The aim of this work is to define a speech emotion recognition (SER) model able to recognize positive, neutral and negative emotions in natural conversations of Italian elderly people. Several datasets for SER are available in the literature. However most of them are in English or Chinese, have been recorded while actors and actresses pronounce short phrases and thus are not related to natural conversation. Moreover only few speeches among all the databases are related to elderly people. Therefore, in this work, a multi-language and multi-age corpus is considered merging a dataset in English, …
Anaphora resolution is one of the major problems in natural language processing. It is also one of the important tasks in machine translation and man/machine dialogue. We solve the problem by using surface expressions and examples. Surface expressions are the words in sentences which provide clues for anaphora resolution. Examples are linguistic data which are actually used in conversations and texts. The method using surface expressions and examples is a practical method. This thesis handles almost all kinds of anaphora: i. The referential property and number of a noun phrase ii. Noun phrase …
This paper concerns how to generate and understand discourse anaphoric noun phrases. I present the results of an analysis of all discourse anaphoric noun phrases (N=1,233) in a corpus of ten narrative monologues, where the choice between a definite pronoun or phrasal NP conforms largely to Gricean constraints on informativeness. I discuss Dale and Reiter's [To appear] recent model and show how it can be augmented for understanding as well as generating the range of data presented here. I argue that integrating centering [Grosz et al., 1983] [Kameyama, 1985] with this model can be applied unifo…
Facebook AI Research（脸书人工智能研究院）；Carnegie Mellon University（卡内基梅隆大学）；Georgia Institute of Technology（佐治亚理工学院）
Visual dialog entails answering a series of questions grounded in an image, using dialog history as context. In addition to the challenges found in visual question answering (VQA), which can be seen as one-round dialog, visual dialog encompasses several more. We focus on one such problem called visual coreference resolution that involves determining which words, typically noun phrases and pronouns, co-refer to the same entity/object instance in an image. This is crucial, especially for pronouns (e.g., `it'), as the dialog agent must first link it to a previous coreference (e.g., `boat'), and o…
The visual dialog task requires an AI agent to interact with humans in multi-round dialogs based on a visual environment. As a common linguistic phenomenon, pronouns are often used in dialogs to improve the communication efficiency. As a result, resolving pronouns (i.e., grounding pronouns to the noun phrases they refer to) is an essential step towards understanding dialogs. In this paper, we propose VD-PCR, a novel framework to improve Visual Dialog understanding with Pronoun Coreference Resolution in both implicit and explicit ways. First, to implicitly help models understand pronouns, we de…
Tsinghua University（清华大学）；The Hong Kong University of Science and Technology（香港科技大学）；Sinovation Ventures（创新工场）
Grounding a pronoun to a visual object it refers to requires complex reasoning from various information sources, especially in conversational scenarios. For example, when people in a conversation talk about something all speakers can see, they often directly use pronouns (e.g., it) to refer to it without previous introduction. This fact brings a huge challenge for modern natural language understanding systems, particularly conventional context-based pronoun coreference models. To tackle this challenge, in this paper, we formally define the task of visual-aware pronoun coreference resolution (P…
Microsoft Research（微软研究院）；UCLA（加州大学洛杉矶分校）；University of Washington（华盛顿大学）；Microsoft Cloud and AI（微软云计算与人工智能部门）；International Digital Economy Academy（国际数字经济研究院）；University of Wisconsin-Madison（威斯康星大学麦迪逊分校）
This paper presents a grounded language-image pre-training (GLIP) model for learning object-level, language-aware, and semantic-rich visual representations. GLIP unifies object detection and phrase grounding for pre-training. The unification brings two benefits: 1) it allows GLIP to learn from both detection and grounding data to improve both tasks and bootstrap a good grounding model; 2) GLIP can leverage massive image-text pairs by generating grounding boxes in a self-training fashion, making the learned representation semantic-rich. In our experiments, we pre-train GLIP on 27M grounding dat…
This paper presents an algorithm for identifying noun-phrase antecedents of pronouns and adjectival anaphors in Spanish dialogues. We believe that anaphora resolution requires numerous sources of information in order to find the correct antecedent of the anaphor. These sources can be of different kinds, e.g., linguistic information, discourse/dialogue structure information, or topic information. For this reason, our algorithm uses various different kinds of information (hybrid information). The algorithm is based on linguistic constraints and preferences and uses an anaphoric accessibility spa…
The contemporary visual captioning models frequently hallucinate objects that are not actually in a scene, due to the visual misclassification or over-reliance on priors that resulting in the semantic inconsistency between the visual information and the target lexical words. The most common way is to encourage the captioning model to dynamically link generated object words or phrases to appropriate regions of the image, i.e., the grounded image captioning (GIC). However, GIC utilizes an auxiliary task (grounding objects) that has not solved the key issue of object hallucination, i.e., the sema…
Several works have proposed to learn a two-path neural network that maps images and texts, respectively, to a same shared Euclidean space where geometry captures useful semantic relationships. Such a multi-modal embedding can be trained and used for various tasks, notably image captioning. In the present work, we introduce a new architecture of this type, with a visual path that leverages recent space-aware pooling mechanisms. Combined with a textual path which is jointly trained from scratch, our semantic-visual embedding offers a versatile model. Once trained under the supervision of caption…
