In this paper, we study the task of image retrieval, where the input query is specified in the form of an image plus some text that describes desired modifications to the input image. For example, we may present an image of the Eiffel tower, and ask the system to find images which are visually similar, but are modified in small ways, such as being taken at nighttime instead of during the day. o tackle this task, we embed the query (reference image plus modification text) and the target (images). The encoding function of the image text query learns a representation, such that the similarity with the target image representation is high iff it is a ``positive match''. We propose a new way to combine image and text through residual connection, that is designed for this retrieval task. We show this outperforms existing approaches on 3 different datasets, namely Fashion-200k, MIT-States and a new synthetic dataset we create based on CLEVR. We also show that our approach can be used to perform image classification with compositionally novel labels, and we outperform previous methods on MIT-States on this task.

The labeling cost of large number of bounding boxes is one of the main challenges for training modern object detectors. To reduce the dependence on expensive bounding box annotations, we propose a new semi-supervised object detection formulation, in which a few seed box level annotations and a large scale of image level annotations are used to train the detector. We adopt a training-mining framework, which is widely used in weakly supervised object detection tasks. However, the mining process inherently introduces various kinds of labelling noises: false negatives, false positives and inaccurate boundaries, which can be harmful for training the standard object detectors (e.g. Faster RCNN). We propose a novel NOise Tolerant Ensemble RCNN (NOTE-RCNN) object detector to handle such noisy labels. Comparing to standard Faster RCNN, it contains three highlights: an ensemble of two classification heads and a distillation head to avoid overfitting on noisy labels and improve the mining precision, masking the negative sample loss in box predictor to avoid the harm of false negative labels, and training box regression head only on seed annotations to eliminate the harm from inaccurate boundaries of mined bounding boxes. We evaluate the methods on ILSVRC 2013 and MSCOCO 2017 dataset; we observe that the detection accuracy consistently improves as we iterate between mining and training steps, and state-of-the-art performance is achieved.

This paper explores object detection in the small data regime, where only a limited number of annotated bounding boxes are available due to data rarity and annotation expense. This is a common challenge today with machine learning being applied to many new tasks where obtaining training data is more challenging, e.g. in medical images with rare diseases that doctors sometimes only see once in their life-time. In this work we explore this problem from a generative modeling perspective by learning to generate new images with associated bounding boxes, and using these for training an object detector. We show that simply training previously proposed generative models does not yield satisfactory performance due to them optimizing for image realism rather than object detection accuracy. To this end we develop a new model with a novel unrolling mechanism that jointly optimizes the generative model and a detector such that the generated images improve the performance of the detector. We show this method outperforms the state of the art on two challenging datasets, disease detection and small data pedestrian detection, improving the average precision on NIH Chest X-ray by a relative 20% and localization accuracy by a relative 50%.

Recent insights on language and vision with neural networks have been successfully applied to simple single-image visual question answering. However, to tackle real-life question answering problems on multimedia collections such as personal photo albums, we have to look at whole collections with sequences of photos. This paper proposes a new multimodal MemexQA task: given a sequence of photos from a user, the goal is to automatically answer questions that help users recover their memory about an event captured in these photos. In addition to a text answer, a few grounding photos are also given to justify the answer. The grounding photos are necessary as they help users quickly verifying the answer. Towards solving the task, we 1) present the MemexQA dataset, the first publicly available multimodal question answering dataset consisting of real personal photo albums; 2) propose an end-to-end trainable network that makes use of a hierarchical process to dynamically determine what media and what time to focus on in the sequential data to answer the question. Experimental results on the MemexQA dataset demonstrate that our model outperforms strong baselines and yields the most relevant grounding photos on this challenging task.

We present a novel framework for iterative visual reasoning. Our framework goes beyond current recognition systems that lack the capability to reason beyond stack of convolutions. The framework consists of two core modules: a local module that uses spatial memory [4] to store previous beliefs with parallel updates; and a global graph-reasoning module. Our graph module has three components: a) a knowledge graph where we represent classes as nodes and build edges to encode different types of semantic relationships between them; b) a region graph of the current image where regions in the image are nodes and spatial relationships between these regions are edges; c) an assignment graph that assigns regions to classes. Both the local module and the global module roll-out iteratively and cross-feed predictions to each other to refine estimates. The final predictions are made by combining the best of both modules with an attention mechanism. We show strong performance over plain ConvNets, e.g. achieving an 8.4% absolute improvement on ADE [55] measured by per-class average precision. Analysis also shows that the framework is resilient to missing regions for reasoning.

Accurate identification and localization of abnormalities from radiology images play an integral part in clinical diagnosis and treatment planning. Building a highly accurate prediction model for these tasks usually requires a large number of images manually annotated with labels and finding sites of abnormalities. In reality, however, such annotated data are expensive to acquire, especially the ones with location annotations. We need methods that can work well with only a small amount of location annotations. To address this challenge, we present a unified approach that simultaneously performs disease identification and localization through the same underlying model for all images. We demonstrate that our approach can effectively leverage both class information as well as limited location annotation, and significantly outperforms the comparative reference baseline in both classification and localization tasks.

Recent insights on language and vision with neural networks have been successfully applied to simple single-image visual question answering. However, to tackle real-life question answering problems on multimedia collections such as personal photos, we have to look at whole collections with sequences of photos or videos. When answering questions from a large collection, a natural problem is to identify snippets to support the answer. In this paper, we describe a novel neural network called Focal Visual-Text Attention network (FVTA) for collective reasoning in visual question answering, where both visual and text sequence information such as images and text metadata are presented. FVTA introduces an end-to-end approach that makes use of a hierarchical process to dynamically determine what media and what time to focus on in the sequential data to answer the question. FVTA can not only answer the questions well but also provides the justifications which the system results are based upon to get the answers. FVTA achieves state-of-the-art performance on the MemexQA dataset and competitive results on the MovieQA dataset.

Recent deep networks are capable of memorizing the entire data even when the labels are completely random. To overcome the overfitting on corrupted labels, we propose a novel technique of learning another neural network, called MentorNet, to supervise the training of the base deep networks, namely, StudentNet. During training, MentorNet provides a curriculum (sample weighting scheme) for StudentNet to focus on the sample the label of which is probably correct. Unlike the existing curriculum that is usually predefined by human experts, MentorNet learns a data-driven curriculum dynamically with StudentNet. Experimental results demonstrate that our approach can significantly improve the generalization performance of deep networks trained on corrupted training data. Notably, to the best of our knowledge, we achieve the best-published result on WebVision, a large benchmark containing 2.2 million images of real-world noisy labels.

One of the most widely used optimization methods for large-scale machine learning problems is distributed asynchronous stochastic gradient descent (DASGD). However, a key issue that arises here is that of delayed gradients: when a “worker” node asynchronously contributes a gradient update to the “master”, the global model parameter may have changed, rendering this information stale. In massively parallel computing grids, these delays can quickly add up if the computational throughput of a node is saturated, so the convergence of DASGD is uncertain under these conditions. Nevertheless, by using a judiciously chosen quasilinear step-size sequence, we show that it is possible to amortize these delays and achieve global convergence with probability 1, even when the delays grow at a polynomial rate. In this way, our results help reaffirm the successful application of DASGD to large-scale optimization problems.

Image captioning is a challenging problem owing to the complexity in understanding the image content and diverse ways of describing it in natural language. Recent advances in deep neural networks have substantially improved the performance of this task. Most state-of-the-art approaches follow an encoder-decoder framework, which generates captions using a sequential recurrent prediction model. However, in this paper, we introduce a novel decision-making framework for image captioning. We utilize a policy network and a value network to collaboratively generate captions. The policy network serves as a local guidance by providing the confidence of predicting the next word according to the current state. Additionally, the value network serves as a global and lookahead guidance by evaluating all possible extensions of the current state. In essence, it adjusts the goal of predicting the correct words towards the goal of generating captions similar to the ground truth captions. We train both networks using an actor-critic reinforcement learning model, with a novel reward defined by visual-semantic embedding. Extensive experiments and analyses on the Microsoft COCO dataset show that the proposed framework outperforms state-of-the-art approaches across different evaluation metrics.

Dense captioning is a newly emerging computer vision topic for understanding images with dense language descriptions. The goal is to densely detect visual concepts (e.g., objects, object parts, and interactions between them) from images, labeling each with a short descriptive phrase. We identify two key challenges of dense captioning that need to be properly addressed when tackling the problem. First, dense visual concept annotations in each image are associated with highly overlapping target regions, making accurate localization of each visual concept challenging. Second, the large amount of visual concepts makes it hard to recognize each of them by appearance alone. We propose a new model pipeline based on two novel ideas, joint inference and context fusion, to alleviate these two challenges. We design our model architecture in a methodical manner and thoroughly evaluate the variations in architecture. Our final model, compact and efficient, achieves state-of-the-art accuracy on Visual Genome [23] for dense captioning with a relative gain of 73% compared to the previous best algorithm. Qualitative experiments also reveal the semantic capabilities of our model in dense captioning.

The ability of learning from noisy labels is very useful in many visual recognition tasks, as a vast amount of data with noisy labels are relatively easy to obtain. Traditionally, label noise has been treated as statistical outliers, and techniques such as importance re-weighting and bootstrapping have been proposed to alleviate the problem. According to our observation, the real-world noisy labels exhibit multimode characteristics as the true labels, rather than behaving like independent random outliers. In this work, we propose a unified distillation framework to use “side” information, including a small clean dataset and label relations in knowledge graph, to “hedge the risk” of learning from noisy labels. Unlike the traditional approaches evaluated based on simulated label noises, we propose a suite of new benchmark datasets, in Sports, Species and Artifacts domains, to evaluate the task of learning from noisy labels in the practical setting. The empirical study demonstrates the effectiveness of our proposed method in all the domains.

Despite progress in perceptual tasks such as image classification, computers still perform poorly on cognitive tasks such as image description and question answering. Cognition is core to tasks that involve not just recognizing, but reasoning about our visual world. However, models used to tackle the rich content in images for cognitive tasks are still being trained using the same datasets designed for perceptual tasks. To achieve success at cognitive tasks, models need to understand the interactions and relationships between objects in an image. When asked “What vehicle is the person riding?”, computers will need to identify the objects in an image as well as the relationships riding(man, carriage) and pulling(horse, carriage) to answer correctly that “the person is riding a horse-drawn carriage.” In this paper, we present the Visual Genome dataset to enable the modeling of such relationships. We collect dense annotations of objects, attributes, and relationships within each image to learn these models. Specifically, our dataset contains over 108K images where each image has an average of $$35$$ objects, $$26$$ attributes, and $$21$$ pairwise relationships between objects. We canonicalize the objects, attributes, relationships, and noun phrases in region descriptions and questions answer pairs to WordNet synsets. Together, these annotations represent the densest and largest dataset of image descriptions, objects, attributes, relationships, and question answer pairs.

There are many cues that can be used to curate media from social networking websites. Beyond metadata, group behavior provide a strong community-based signal for surfacing images, which we show in a user-defined curatorial task. In a departure from mirco-task crowdwork, we observe that the curation inherent in online photo communities guides the discoverability and consumption of the media, which in turn provides a strong signal that can be used in new editorial tasks in a community-supervised manner. We use this approach in tandem with other more conventional multimedia methods (i.e.\ computer vision and contextual metadata) to form a broad multimodal approach to retrieval and recommendation. We present a large-scale system implementation on a real-world curative task for weather images on a web-scale dataset. Finally, we conduct an evaluation of this system using professional editors and find substantial improvements in editorial efficiency.

This paper develops a novel framework for semantic image retrieval based on the notion of a scene graph. Our scene graphs represent objects (“man”, “boat”), attributes of objects (“boat is white”) and relationships between objects (“man standing on boat”). We use these scene graphs as queries to retrieve semantically related images. To this end, we design a conditional random field model that reasons about possible groundings of scene graphs to test images. The likelihoods of these groundings are used as ranking scores for retrieval. We introduce a novel dataset of 5,000 human-generated scene graphs grounded to images and use this dataset to evaluate our method for image retrieval. In particular, we evaluate retrieval using full scene graphs and small scene subgraphs, and show that our method outperforms retrieval methods that use only objects or low-level image features. In addition, we show that our full model can be used to improve object localization compared to baseline methods.

The long-standing goal of localizing every object in an image remains elusive. Manually annotating objects is quite expensive despite crowd engineering innovations. Current state-of-the-art automatic object detectors can accurately detect at most a few objects per image. This paper brings together the latest advancements in object detection and in crowd engineering into a principled framework for accurately and efficiently localizing objects in images. The input to the system is an image to annotate and a set of annotation constraints: desired precision, utility and/or human cost of the labeling. The output is a set of object annotations, informed by human feedback and computer vision. Our model seamlessly integrates multiple computer vision models with multiple sources of human input in a Markov Decision Process. We empirically validate the effectiveness of our human-in-the-loop labeling approach on the ILSVRC2014 object detection dataset.

In this paper we consider the problem of multi-view face detection. While there has been significant research on this problem, current state-of-the-art approaches for this task require annotation of facial landmarks, e.g. TSM [25], or annotation of face poses [28, 22]. They also require training dozens of models to fully capture faces in all orientations, e.g. 22 models in HeadHunter method [22]. In this paper we propose Deep Dense Face Detector (DDFD), a method that does not require pose/landmark annotation and is able to detect faces in a wide range of orientations using a single model based on deep convolutional neural networks. The proposed method has minimal complexity; unlike other recent deep learning object detection methods [9], it does not require additional components such as segmentation, bounding-box regression, or SVM classifiers. Furthermore, we analyzed scores of the proposed face detector for faces in different orientations and found that 1) the proposed method is able to detect faces from different angles and can handle occlusion to some extent, 2) there seems to be a correlation between distribution of positive examples in the training set and scores of the proposed face detector. The latter suggests that the proposed method's performance can be further improved by using better sampling strategies and more sophisticated data augmentation techniques. Evaluations on popular face detection benchmark datasets show that our single-model face detector algorithm has similar or better performance compared to the previous methods, which are more complex and require annotations of either different poses or facial landmarks.

This ACM MM panel aims to redefine the state of research between Academia and Industry.

Photos are an important information carrier for implicit relationships. In this article, we introduce an image based social network, called CelebrityNet , built from implicit relationships encoded in a collection of celebrity images. We analyze the social properties reflected in this image-based social network and automatically infer communities among the celebrities. We demonstrate the interesting discoveries of the CelebrityNet. We particularly compare the inferred communities with human manually labeled ones and show quantitatively that the automatically detected communities are highly aligned with that of human interpretation. Inspired by the uniqueness of visual content and tag concepts within each community of the CelebrityNet, we further demonstrate that the constructed social network can serve as a knowledge base for high-level visual recognition tasks. In particular, this social network is capable of significantly improving the performance of automatic image annotation and classification of unknown images.

In this paper, we tackle the problem of co-localization in real-world images. Co-localization is the problem of simultaneously localizing (with bounding boxes) objects of the same class across a set of distinct images. Although similar problems such as co-segmentation and weakly supervised localization have been previously studied, we focus on being able to perform co-localization in real-world settings, which are typically characterized by large amounts of intra-class variation, inter-class diversity, and annotation noise. To address these issues, we present a joint image-box formulation for solving the co-localization problem, and show how it can be relaxed to a convex quadratic program which can be efficiently solved. We perform an extensive evaluation of our method compared to previous state-of-the-art approaches on the challenging PASCAL VOC 2007 and Object Discovery datasets. In addition, we also present a large-scale study of co-localization on ImageNet, involving ground-truth annotations for 3, 624 classes and approximately 1 million images.

With the rapid development of e-commerce and explosive growth of online shopping market, the problem of how to optimize the process of guiding the user to the huge amount of online products have been urgent. Existing recommender systems use information from users' profiles (demographic filtering), similar neighbors (collaborative filtering), and textual description (content-based model) to make recommendations, which easily generate irrelevant suggestions to users due to the ignorance of users' intentions and the visual similarity among products. In this paper, we proposed an interactive product recommendation method, which considers not only the product diversity but also the visual similarity, to interactively capture a user's real intention and refine the product recommendation result based on the user's real product interests. Our algorithm is experimentally evaluated under a real-world user log, and shown to significantly improve recommendation accuracy over the traditional approaches.

This paper addresses the problem of fine-grained recognition: recognizing subordinate categories such as bird species, car models, or dog breeds. We focus on two major challenges: learning expressive appearance descriptors and localizing discriminative parts. To this end, we propose an object representation that detects important parts and describes fine grained appearances. The part detectors are learned in a fully unsupervised manner, based on the insight that images with similar poses can be automatically discovered for fine-grained classes in the same domain. The appearance descriptors are learned using a convolutional neural network. Our approach requires only image level class labels, without any use of part annotations or segmentation masks, which may be costly to obtain. We show experimentally that combining these two insights is an effective strategy for fine-grained recognition.

Multi-label learning deals with the classification problems where each instance can be assigned with multiple labels simultaneously. Conventional multi-label learning approaches mainly focus on exploiting label correlations. It is usually assumed, explicitly or implicitly, that the label sets for training instances are fully labeled without any missing labels. However, in many real-world multi-label datasets, the label assignments for training instances can be incomplete. Some ground-truth labels can be missed by the labeler from the label set. This problem is especially typical when the number instances is very large, and the labeling cost is very high, which makes it almost impossible to get a fully labeled training set. In this paper, we study the problem of large-scale multi-label learning with incomplete label assignments. We propose an approach, called Mpu, based upon positive and unlabeled stochastic gradient descent and stacked models. Unlike prior works, our method can effectively and efficiently consider missing labels and label correlations simultaneously, and is very scalable, that has linear time complexities over the size of the data. Extensive experiments on two real-world multi-label datasets show that our Mpu model consistently outperform other commonly-used baselines.

We propose a new approach for constructing mid-level visual features for image classification. We represent an image using the outputs of a collection of binary classifiers. These binary classifiers are trained to differentiate pairs of object classes in an object hierarchy. Our feature representation implicitly captures the hierarchical structure in object classes. We show that our proposed approach outperforms other baseline methods in image classification.

We present a scalable approach to automatically suggest relevant clothing products, given a single image without metadata. We formulate the problem as cross-scenario retrieval: the query is a real-world image, while the products from online shopping catalogs are usually presented in a clean environment. We divide our approach into two main stages: a) Starting from articulated pose estimation, we segment the person area and cluster promising image regions in order to detect the clothing classes present in the query image. b) We use image retrieval techniques to retrieve visually similar products from each of the detected classes. We achieve clothing detection performance comparable to the state-of-the-art on a very recent annotated dataset, while being more than 50 times faster. Finally, we present a large scale clothing suggestion scenario, where the product database contains over one million products.

A semantically meaningful image hierarchy can ease the human effort in organizing thousands and millions of pictures (e.g., personal albums), and help to improve performance of end tasks such as image annotation and classification. Previous work has focused on using either low-level image features or textual tags to build image hierarchies, resulting in limited success in their general usage. In this paper, we propose a method to automatically discover the “semantivisual” image hierarchy by incorporating both image and tag information. This hierarchy encodes a general-to-specific image relationship. We pay particular attention to quantifying the effectiveness of the learned hierarchy, as well as comparing our method with others in the end-task applications. Our experiments show that humans find our semantivisual image hierarchy more effective than those solely based on texts or low-level visual features. And using the constructed image hierarchy as a knowledge ontology, our algorithm can perform challenging image classification and annotation tasks more accurately.

Robust low-level image features have been proven to be effective representations for a variety of visual recognition tasks such as object recognition and scene classification; but pixels, or even local image patches, carry little semantic meanings. For high level visual tasks, such low-level image representations are potentially not enough. In this paper, we propose a high-level image representation, called the Object Bank, where an image is represented as a scale invariant response map of a large number of pre-trained generic object detectors, blind to the testing dataset or visual task. Leveraging on the Object Bank representation, superior performances on high level visual recognition tasks can be achieved with simple off-the-shelf classifiers such as logistic regression and linear SVM. Sparsity algorithms make our representation more efficient and scalable for large scene datasets, and reveal semantically meaningful feature patterns.

Upstream supervised topic models have been widely used for complicated scene understanding. However, existing maximum likelihood estimation (MLE) schemes can make the prediction model learning independent of latent topic discovery and result in an imbalanced prediction rule for scene classification. This paper presents a joint max-margin and max-likelihood learning method for upstream scene understanding models, in which latent topic discovery and prediction model estimation are closely coupled and well-balanced. The optimization problem is efficiently solved with a variational EM procedure, which iteratively solves an online loss-augmented SVM. We demonstrate the advantages of the large-margin approach on both an 8-category sports dataset and the 67-class MIT indoor scene dataset for scene categorization.

The explosion of image data on the Internet has the potential to foster more sophisticated and robust models and algorithms to index, retrieve, organize and interact with images and multimedia data. But exactly how such data can be harnessed and organized remains a critical problem. We introduce here a new database called “ImageNet”, a large-scale ontology of images built upon the backbone of the WordNet structure. ImageNet aims to populate the majority of the 80,000 synsets of WordNet with an average of 500–1000 clean and full resolution images. This will result in tens of millions of annotated images organized by the semantic hierarchy of WordNet. This paper offers a detailed analysis of ImageNet in its current state: 12 subtrees with 5247 synsets and 3.2 million images in total. We show that ImageNet is much larger in scale and diversity and much more accurate than the current image datasets. Constructing such a large-scale database is a challenging task. We describe the data collection scheme with Amazon Mechanical Turk. Lastly, we illustrate the usefulness of ImageNet through three simple applications in object recognition, image classification and automatic object clustering. We hope that the scale, accuracy, diversity and hierarchical structure of ImageNet can offer unparalleled opportunities to researchers in the computer vision community and beyond.

Given an image, we propose a hierarchical generative model that classifies the overall scene, recognizes and segments each object component, as well as annotates the image with a list of tags. To our knowledge, this is the first model that performs all three tasks in one coherent framework. For instance, a scene of a `polo game' consists of several visual objects such as `human', `horse', `grass', etc. In addition, it can be further annotated with a list of more abstract (e.g. `dusk') or visually less salient (e.g. `saddle') tags. Our generative model jointly explains images through a visual model and a textual model. Visually relevant objects are represented by regions and patches, while visually irrelevant textual annotations are influenced directly by the overall scene class. We propose a fully automatic learning framework that is able to learn robust scene models from noisy Web data such as images and user tags from Flickr.com. We demonstrate the effectiveness of our framework by automatically classifying, annotating and segmenting images from eight classes depicting sport scenes. In all three tasks, our model significantly outperforms state-of-the-art algorithms.

A well-built dataset is a necessary starting point for advanced computer vision research. It plays a crucial role in evaluation and provides a continuous challenge to state-of-the-art algorithms. Dataset collection is, however, a tedious and time-consuming task. This paper presents a novel automatic dataset collecting and model learning approach that uses object recognition techniques in an incremental method. The goal of this work is to use the tremendous resources of the web to learn robust object category models in order to detect and search for objects in real-world cluttered scenes. It mimics the human learning process of iteratively accumulating model knowledge and image examples. We adapt a non-parametric graphical model and propose an incremental learning framework. Our algorithm is capable of automatically collecting much larger object category datasets for 22 randomly selected classes from the Caltech 101 dataset. Furthermore, we offer not only more images in each object category dataset, but also a robust object model and meaningful image annotation. Our experiments show that OPTIMOL is capable of collecting image datasets that are superior to Caltech 101 and LabelMe.

We propose a first attempt to classify events in static images by integrating scene and object categorizations. We define an event in a static image as a human activity taking place in a specific environment. In this paper, we use a number of sport games such as snow boarding, rock climbing or badminton to demonstrate event classification. Our goal is to classify the event in the image as well as to provide a number of semantic labels to the objects and scene environment within the image. For example, given a rowing scene, our algorithm recognizes the event as rowing by classifying the environment as a lake and recognizing the critical objects in the image as athletes, rowing boat, water, etc. We achieve this integrative and holistic recognition through a generative graphical model. We have assembled a highly challenging database of 8 widely varied sport events. We show that our system is capable of classifying these event classes at 73.4% accuracy. While each component of the model contributes to the final recognition, using scene or objects alone cannot achieve this performance.