Shih-Fu Chang

Zhecan Wang, Rui Sun, Haoxuan You et al.

Vision-language tasks, such as VQA, SNLI-VE, and VCR are challenging because they require the model's reasoning ability to understand the semantics of the visual world and natural language. Supervised methods working for vision-language tasks have been well-studied. However, solving these tasks in a zero-shot setting is less explored. Since Contrastive Language-Image Pre-training (CLIP) has shown remarkable zero-shot performance on image-text matching, previous works utilized its strong zero-shot ability by converting vision-language tasks into an image-text matching problem, and they mainly consider global-level matching (e.g., the whole image or sentence). However, we find visual and textual fine-grained information, e.g., keywords in the sentence and objects in the image, can be fairly informative for semantics understanding. Inspired by this, we propose a unified framework to take advantage of the fine-grained information for zero-shot vision-language learning, covering multiple tasks such as VQA, SNLI-VE, and VCR. Our experiments show that our framework outperforms former zero-shot methods on VQA and achieves substantial improvement on SNLI-VE and VCR. Furthermore, our ablation studies confirm the effectiveness and generalizability of our proposed method.

Zhecan Wang, Long Chen, Haoxuan You et al.

Vision-language (VL) understanding tasks evaluate models' comprehension of complex visual scenes through multiple-choice questions. However, we have identified two dataset biases that models can exploit as shortcuts to resolve various VL tasks correctly without proper understanding. The first type of dataset bias is \emph{Unbalanced Matching} bias, where the correct answer overlaps the question and image more than the incorrect answers. The second type of dataset bias is \emph{Distractor Similarity} bias, where incorrect answers are overly dissimilar to the correct answer but significantly similar to other incorrect answers within the same sample. To address these dataset biases, we first propose Adversarial Data Synthesis (ADS) to generate synthetic training and debiased evaluation data. We then introduce Intra-sample Counterfactual Training (ICT) to assist models in utilizing the synthesized training data, particularly the counterfactual data, via focusing on intra-sample differentiation. Extensive experiments demonstrate the effectiveness of ADS and ICT in consistently improving model performance across different benchmarks, even in domain-shifted scenarios.

Haoxuan You, Zhecan Wang, Rui Sun et al.

The field of vision-and-language (VL) understanding has made unprecedented progress with end-to-end large pre-trained VL models (VLMs). However, they still fall short in zero-shot reasoning tasks that require multi-step inferencing. To achieve this goal, previous works resort to a divide-and-conquer pipeline. In this paper, we argue that previous efforts have several inherent shortcomings: 1) They rely on domain-specific sub-question decomposing models. 2) They force models to predict the final answer even if the sub-questions or sub-answers provide insufficient information. We address these limitations via IdealGPT, a framework that iteratively decomposes VL reasoning using large language models (LLMs). Specifically, IdealGPT utilizes an LLM to generate sub-questions, a VLM to provide corresponding sub-answers, and another LLM to reason to achieve the final answer. These three modules perform the divide-and-conquer procedure iteratively until the model is confident about the final answer to the main question. We evaluate IdealGPT on multiple challenging VL reasoning tasks under a zero-shot setting. In particular, our IdealGPT outperforms the best existing GPT-4-like models by an absolute 10% on VCR and 15% on SNLI-VE. Code is available at https://github.com/Hxyou/IdealGPT

Hassan Akbari, Hamid Palangi, Jianwei Yang et al.

Neuro-symbolic representations have proved effective in learning structure information in vision and language. In this paper, we propose a new model architecture for learning multi-modal neuro-symbolic representations for video captioning. Our approach uses a dictionary learning-based method of learning relations between videos and their paired text descriptions. We refer to these relations as relative roles and leverage them to make each token role-aware using attention. This results in a more structured and interpretable architecture that incorporates modality-specific inductive biases for the captioning task. Intuitively, the model is able to learn spatial, temporal, and cross-modal relations in a given pair of video and text. The disentanglement achieved by our proposal gives the model more capacity to capture multi-modal structures which result in captions with higher quality for videos. Our experiments on two established video captioning datasets verifies the effectiveness of the proposed approach based on automatic metrics. We further conduct a human evaluation to measure the grounding and relevance of the generated captions and observe consistent improvement for the proposed model. The codes and trained models can be found at https://github.com/hassanhub/R3Transformer

Hammad Ayyubi, Junzhang Liu, Ali Asgarov et al.

In this paper, we present ENTER, an interpretable Video Question Answering (VideoQA) system based on event graphs. Event graphs convert videos into graphical representations, where video events form the nodes and event-event relationships (temporal/causal/hierarchical) form the edges. This structured representation offers many benefits: 1) Interpretable VideoQA via generated code that parses event-graph; 2) Incorporation of contextual visual information in the reasoning process (code generation) via event graphs; 3) Robust VideoQA via Hierarchical Iterative Update of the event graphs. Existing interpretable VideoQA systems are often top-down, disregarding low-level visual information in the reasoning plan generation, and are brittle. While bottom-up approaches produce responses from visual data, they lack interpretability. Experimental results on NExT-QA, IntentQA, and EgoSchema demonstrate that not only does our method outperform existing top-down approaches while obtaining competitive performance against bottom-up approaches, but more importantly, offers superior interpretability and explainability in the reasoning process.

Hammad Ayyubi, Xuande Feng, Junzhang Liu et al.

The task of predicting time and location from images is challenging and requires complex human-like puzzle-solving ability over different clues. In this work, we formalize this ability into core skills and implement them using different modules in an expert pipeline called PuzzleGPT. PuzzleGPT consists of a perceiver to identify visual clues, a reasoner to deduce prediction candidates, a combiner to combinatorially combine information from different clues, a web retriever to get external knowledge if the task can't be solved locally, and a noise filter for robustness. This results in a zero-shot, interpretable, and robust approach that records state-of-the-art performance on two datasets -- TARA and WikiTilo. PuzzleGPT outperforms large VLMs such as BLIP-2, InstructBLIP, LLaVA, and even GPT-4V, as well as automatically generated reasoning pipelines like VisProg, by at least 32% and 38%, respectively. It even rivals or surpasses finetuned models.

Brian Chen, Xudong Lin, Christopher Thomas et al.

Visual and textual modalities contribute complementary information about events described in multimedia documents. Videos contain rich dynamics and detailed unfoldings of events, while text describes more high-level and abstract concepts. However, existing event extraction methods either do not handle video or solely target video while ignoring other modalities. In contrast, we propose the first approach to jointly extract events from video and text articles. We introduce the new task of Video MultiMedia Event Extraction (Video M2E2) and propose two novel components to build the first system towards this task. First, we propose the first self-supervised multimodal event coreference model that can determine coreference between video events and text events without any manually annotated pairs. Second, we introduce the first multimodal transformer which extracts structured event information jointly from both videos and text documents. We also construct and will publicly release a new benchmark of video-article pairs, consisting of 860 video-article pairs with extensive annotations for evaluating methods on this task. Our experimental results demonstrate the effectiveness of our proposed method on our new benchmark dataset. We achieve 6.0% and 5.8% absolute F-score gain on multimodal event coreference resolution and multimedia event extraction.

Jiawei Ma, Hanchen Xie, Guangxing Han et al.

Few-shot Learning has been studied to mimic human visual capabilities and learn effective models without the need of exhaustive human annotation. Even though the idea of meta-learning for adaptation has dominated the few-shot learning methods, how to train a feature extractor is still a challenge. In this paper, we focus on the design of training strategy to obtain an elemental representation such that the prototype of each novel class can be estimated from a few labeled samples. We propose a two-stage training scheme, Partner-Assisted Learning (PAL), which first trains a partner encoder to model pair-wise similarities and extract features serving as soft-anchors, and then trains a main encoder by aligning its outputs with soft-anchors while attempting to maximize classification performance. Two alignment constraints from logit-level and feature-level are designed individually. For each few-shot task, we perform prototype classification. Our method consistently outperforms the state-of-the-art method on four benchmarks. Detailed ablation studies of PAL are provided to justify the selection of each component involved in training.

Brian Chen, Andrew Rouditchenko, Kevin Duarte et al.

Multimodal self-supervised learning is getting more and more attention as it allows not only to train large networks without human supervision but also to search and retrieve data across various modalities. In this context, this paper proposes a self-supervised training framework that learns a common multimodal embedding space that, in addition to sharing representations across different modalities, enforces a grouping of semantically similar instances. To this end, we extend the concept of instance-level contrastive learning with a multimodal clustering step in the training pipeline to capture semantic similarities across modalities. The resulting embedding space enables retrieval of samples across all modalities, even from unseen datasets and different domains. To evaluate our approach, we train our model on the HowTo100M dataset and evaluate its zero-shot retrieval capabilities in two challenging domains, namely text-to-video retrieval, and temporal action localization, showing state-of-the-art results on four different datasets.

Liunian Harold Li, Haoxuan You, Zhecan Wang et al.

Pre-trained contextual vision-and-language (V&L) models have achieved impressive performance on various benchmarks. However, existing models require a large amount of parallel image-caption data for pre-training. Such data are costly to collect and require cumbersome curation. Inspired by unsupervised machine translation, we investigate if a strong V&L representation model can be learned through unsupervised pre-training without image-caption corpora. In particular, we propose to conduct ``mask-and-predict'' pre-training on text-only and image-only corpora and introduce the object tags detected by an object recognition model as anchor points to bridge two modalities. We find that such a simple approach achieves performance close to a model pre-trained with aligned data, on four English V&L benchmarks. Our work challenges the widely held notion that aligned data is necessary for V&L pre-training, while significantly reducing the amount of supervision needed for V&L models.

Bo Wu, Haoyu Qin, Alireza Zareian et al.

Children acquire language subconsciously by observing the surrounding world and listening to descriptions. They can discover the meaning of words even without explicit language knowledge, and generalize to novel compositions effortlessly. In this paper, we bring this ability to AI, by studying the task of Visually grounded Language Acquisition (VLA). We propose a multimodal transformer model augmented with a novel mechanism for analogical reasoning, which approximates novel compositions by learning semantic mapping and reasoning operations from previously seen compositions. Our proposed method, Analogical Reasoning Transformer Networks (ARTNet), is trained on raw multimedia data (video frames and transcripts), and after observing a set of compositions such as "washing apple" or "cutting carrot", it can generalize and recognize new compositions in new video frames, such as "washing carrot" or "cutting apple". To this end, ARTNet refers to relevant instances in the training data and uses their visual features and captions to establish analogies with the query image. Then it chooses the suitable verb and noun to create a new composition that describes the new image best. Extensive experiments on an instructional video dataset demonstrate that the proposed method achieves significantly better generalization capability and recognition accuracy compared to state-of-the-art transformer models.

Qingyun Wang, Manling Li, Xuan Wang et al.

To combat COVID-19, both clinicians and scientists need to digest vast amounts of relevant biomedical knowledge in scientific literature to understand the disease mechanism and related biological functions. We have developed a novel and comprehensive knowledge discovery framework, COVID-KG to extract fine-grained multimedia knowledge elements (entities and their visual chemical structures, relations, and events) from scientific literature. We then exploit the constructed multimedia knowledge graphs (KGs) for question answering and report generation, using drug repurposing as a case study. Our framework also provides detailed contextual sentences, subfigures, and knowledge subgraphs as evidence.

Alireza Zareian, Zhecan Wang, Haoxuan You et al.

Scene graph generation models understand the scene through object and predicate recognition, but are prone to mistakes due to the challenges of perception in the wild. Perception errors often lead to nonsensical compositions in the output scene graph, which do not follow real-world rules and patterns, and can be corrected using commonsense knowledge. We propose the first method to acquire visual commonsense such as affordance and intuitive physics automatically from data, and use that to improve the robustness of scene understanding. To this end, we extend Transformer models to incorporate the structure of scene graphs, and train our Global-Local Attention Transformer on a scene graph corpus. Once trained, our model can be applied on any scene graph generation model and correct its obvious mistakes, resulting in more semantically plausible scene graphs. Through extensive experiments, we show our model learns commonsense better than any alternative, and improves the accuracy of state-of-the-art scene graph generation methods.

Zhizheng Zhang, Cuiling Lan, Wenjun Zeng et al.

Person Re-identification (ReID) aims at matching a person of interest across images. In convolutional neural network (CNN) based approaches, loss design plays a vital role in pulling closer features of the same identity and pushing far apart features of different identities. In recent years, triplet loss achieves superior performance and is predominant in ReID. However, triplet loss considers only three instances of two classes in per-query optimization (with an anchor sample as query) and it is actually equivalent to a two-class classification. There is a lack of loss design which enables the joint optimization of multiple instances (of multiple classes) within per-query optimization for person ReID. In this paper, we introduce a multi-class classification loss, i.e., N-tuple loss, to jointly consider multiple (N) instances for per-query optimization. This in fact aligns better with the ReID test/inference process, which conducts the ranking/comparisons among multiple instances. Furthermore, for more efficient multi-class classification, we propose a new meta prototypical N-tuple loss. With the multi-class classification incorporated, our model achieves the state-of-the-art performance on the benchmark person ReID datasets.

Manling Li, Alireza Zareian, Qi Zeng et al.

We introduce a new task, MultiMedia Event Extraction (M2E2), which aims to extract events and their arguments from multimedia documents. We develop the first benchmark and collect a dataset of 245 multimedia news articles with extensively annotated events and arguments. We propose a novel method, Weakly Aligned Structured Embedding (WASE), that encodes structured representations of semantic information from textual and visual data into a common embedding space. The structures are aligned across modalities by employing a weakly supervised training strategy, which enables exploiting available resources without explicit cross-media annotation. Compared to uni-modal state-of-the-art methods, our approach achieves 4.0% and 9.8% absolute F-score gains on text event argument role labeling and visual event extraction. Compared to state-of-the-art multimedia unstructured representations, we achieve 8.3% and 5.0% absolute F-score gains on multimedia event extraction and argument role labeling, respectively. By utilizing images, we extract 21.4% more event mentions than traditional text-only methods.

Yang Feng, Futang Peng, Xu Zhang et al.

Deep image embedding provides a way to measure the semantic similarity of two images. It plays a central role in many applications such as image search, face verification, and zero-shot learning. It is desirable to have a universal deep embedding model applicable to various domains of images. However, existing methods mainly rely on training specialist embedding models each of which is applicable to images from a single domain. In this paper, we study an important but unexplored task: how to train a single universal image embedding model to match the performance of several specialists on each specialist's domain. Simply fusing the training data from multiple domains cannot solve this problem because some domains become overfitted sooner when trained together using existing methods. Therefore, we propose to distill the knowledge in multiple specialists into a universal embedding to solve this problem. In contrast to existing embedding distillation methods that distill the absolute distances between images, we transform the absolute distances between images into a probabilistic distribution and minimize the KL-divergence between the distributions of the specialists and the universal embedding. Using several public datasets, we validate that our proposed method accomplishes the goal of universal image embedding.

Alireza Zareian, Svebor Karaman, Shih-Fu Chang

Scene Graph Generation (SGG) aims to extract entities, predicates and their semantic structure from images, enabling deep understanding of visual content, with many applications such as visual reasoning and image retrieval. Nevertheless, existing SGG methods require millions of manually annotated bounding boxes for training, and are computationally inefficient, as they exhaustively process all pairs of object proposals to detect predicates. In this paper, we address those two limitations by first proposing a generalized formulation of SGG, namely Visual Semantic Parsing, which disentangles entity and predicate recognition, and enables sub-quadratic performance. Then we propose the Visual Semantic Parsing Network, VSPNet, based on a dynamic, attention-based, bipartite message passing framework that jointly infers graph nodes and edges through an iterative process. Additionally, we propose the first graph-based weakly supervised learning framework, based on a novel graph alignment algorithm, which enables training without bounding box annotations. Through extensive experiments, we show that VSPNet outperforms weakly supervised baselines significantly and approaches fully supervised performance, while being several times faster. We publicly release the source code of our method.

Alireza Zareian, Svebor Karaman, Shih-Fu Chang

Scene graphs are powerful representations that parse images into their abstract semantic elements, i.e., objects and their interactions, which facilitates visual comprehension and explainable reasoning. On the other hand, commonsense knowledge graphs are rich repositories that encode how the world is structured, and how general concepts interact. In this paper, we present a unified formulation of these two constructs, where a scene graph is seen as an image-conditioned instantiation of a commonsense knowledge graph. Based on this new perspective, we re-formulate scene graph generation as the inference of a bridge between the scene and commonsense graphs, where each entity or predicate instance in the scene graph has to be linked to its corresponding entity or predicate class in the commonsense graph. To this end, we propose a novel graph-based neural network that iteratively propagates information between the two graphs, as well as within each of them, while gradually refining their bridge in each iteration. Our Graph Bridging Network, GB-Net, successively infers edges and nodes, allowing to simultaneously exploit and refine the rich, heterogeneous structure of the interconnected scene and commonsense graphs. Through extensive experimentation, we showcase the superior accuracy of GB-Net compared to the most recent methods, resulting in a new state of the art. We publicly release the source code of our method.

Jiawei Ma, Zheng Shou, Alireza Zareian et al.

Many real-world applications involve multivariate, geo-tagged time series data: at each location, multiple sensors record corresponding measurements. For example, air quality monitoring system records PM2.5, CO, etc. The resulting time-series data often has missing values due to device outages or communication errors. In order to impute the missing values, state-of-the-art methods are built on Recurrent Neural Networks (RNN), which process each time stamp sequentially, prohibiting the direct modeling of the relationship between distant time stamps. Recently, the self-attention mechanism has been proposed for sequence modeling tasks such as machine translation, significantly outperforming RNN because the relationship between each two time stamps can be modeled explicitly. In this paper, we are the first to adapt the self-attention mechanism for multivariate, geo-tagged time series data. In order to jointly capture the self-attention across multiple dimensions, including time, location and the sensor measurements, while maintain low computational complexity, we propose a novel approach called Cross-Dimensional Self-Attention (CDSA) to process each dimension sequentially, yet in an order-independent manner. Our extensive experiments on four real-world datasets, including three standard benchmarks and our newly collected NYC-traffic dataset, demonstrate that our approach outperforms the state-of-the-art imputation and forecasting methods. A detailed systematic analysis confirms the effectiveness of our design choices.

Xu Zhang, Felix Xinnan Yu, Svebor Karaman et al.

Metric learning aims at learning a distance which is consistent with the semantic meaning of the samples. The problem is generally solved by learning an embedding for each sample such that the embeddings of samples of the same category are compact while the embeddings of samples of different categories are spread-out in the feature space. We study the features extracted from the second last layer of a deep neural network based classifier trained with the cross entropy loss on top of the softmax layer. We show that training classifiers with different temperature values of softmax function leads to features with different levels of compactness. Leveraging these insights, we propose a "heating-up" strategy to train a classifier with increasing temperatures, leading the corresponding embeddings to achieve state-of-the-art performance on a variety of metric learning benchmarks.

Philipp Blandfort, Desmond Patton, William R. Frey et al.

Gang violence is a severe issue in major cities across the U.S. and recent studies [Patton et al. 2017] have found evidence of social media communications that can be linked to such violence in communities with high rates of exposure to gang activity. In this paper we partnered computer scientists with social work researchers, who have domain expertise in gang violence, to analyze how public tweets with images posted by youth who mention gang associations on Twitter can be leveraged to automatically detect psychosocial factors and conditions that could potentially assist social workers and violence outreach workers in prevention and early intervention programs. To this end, we developed a rigorous methodology for collecting and annotating tweets. We gathered 1,851 tweets and accompanying annotations related to visual concepts and the psychosocial codes: aggression, loss, and substance use. These codes are relevant to social work interventions, as they represent possible pathways to violence on social media. We compare various methods for classifying tweets into these three classes, using only the text of the tweet, only the image of the tweet, or both modalities as input to the classifier. In particular, we analyze the usefulness of mid-level visual concepts and the role of different modalities for this tweet classification task. Our experiments show that individually, text information dominates classification performance of the loss class, while image information dominates the aggression and substance use classes. Our multimodal approach provides a very promising improvement (18% relative in mean average precision) over the best single modality approach. Finally, we also illustrate the complexity of understanding social media data and elaborate on open challenges.

Di Lu, Spencer Whitehead, Lifu Huang et al.

Current image captioning approaches generate descriptions which lack specific information, such as named entities that are involved in the images. In this paper we propose a new task which aims to generate informative image captions, given images and hashtags as input. We propose a simple but effective approach to tackle this problem. We first train a convolutional neural networks - long short term memory networks (CNN-LSTM) model to generate a template caption based on the input image. Then we use a knowledge graph based collective inference algorithm to fill in the template with specific named entities retrieved via the hashtags. Experiments on a new benchmark dataset collected from Flickr show that our model generates news-style image descriptions with much richer information. Our model outperforms unimodal baselines significantly with various evaluation metrics.

Victor Campos, Brendan Jou, Xavier Giro-i-Nieto et al.

Recurrent Neural Networks (RNNs) continue to show outstanding performance in sequence modeling tasks. However, training RNNs on long sequences often face challenges like slow inference, vanishing gradients and difficulty in capturing long term dependencies. In backpropagation through time settings, these issues are tightly coupled with the large, sequential computational graph resulting from unfolding the RNN in time. We introduce the Skip RNN model which extends existing RNN models by learning to skip state updates and shortens the effective size of the computational graph. This model can also be encouraged to perform fewer state updates through a budget constraint. We evaluate the proposed model on various tasks and show how it can reduce the number of required RNN updates while preserving, and sometimes even improving, the performance of the baseline RNN models. Source code is publicly available at https://imatge-upc.github.io/skiprnn-2017-telecombcn/ .

Hanwang Zhang, Zawlin Kyaw, Jinyang Yu et al.

We aim to tackle a novel vision task called Weakly Supervised Visual Relation Detection (WSVRD) to detect "subject-predicate-object" relations in an image with object relation groundtruths available only at the image level. This is motivated by the fact that it is extremely expensive to label the combinatorial relations between objects at the instance level. Compared to the extensively studied problem, Weakly Supervised Object Detection (WSOD), WSVRD is more challenging as it needs to examine a large set of regions pairs, which is computationally prohibitive and more likely stuck in a local optimal solution such as those involving wrong spatial context. To this end, we present a Parallel, Pairwise Region-based, Fully Convolutional Network (PPR-FCN) for WSVRD. It uses a parallel FCN architecture that simultaneously performs pair selection and classification of single regions and region pairs for object and relation detection, while sharing almost all computation shared over the entire image. In particular, we propose a novel position-role-sensitive score map with pairwise RoI pooling to efficiently capture the crucial context associated with a pair of objects. We demonstrate the superiority of PPR-FCN over all baselines in solving the WSVRD challenge by using results of extensive experiments over two visual relation benchmarks.

Hongzhi Li, Joseph G. Ellis, Shih-Fu Chang

In this paper we describe a novel framework and algorithms for discovering image patch patterns from a large corpus of weakly supervised image-caption pairs generated from news events. Current pattern mining techniques attempt to find patterns that are representative and discriminative, we stipulate that our discovered patterns must also be recognizable by humans and preferably with meaningful names. We propose a new multimodal pattern mining approach that leverages the descriptive captions often accompanying news images to learn semantically meaningful image patch patterns. The mutltimodal patterns are then named using words mined from the associated image captions for each pattern. A novel evaluation framework is provided that demonstrates our patterns are 26.2% more semantically meaningful than those discovered by the state of the art vision only pipeline, and that we can provide tags for the discovered images patches with 54.5% accuracy with no direct supervision. Our methods also discover named patterns beyond those covered by the existing image datasets like ImageNet. To the best of our knowledge this is the first algorithm developed to automatically mine image patch patterns that have strong semantic meaning specific to high-level news events, and then evaluate these patterns based on that criteria.