Human visual attention is a complex phenomenon. A computational modeling of this phenomenon must take into account where people look in order to evaluate which are the salient locations (spatial distribution of the fixations), when they look in those locations to understand the temporal development of the exploration (temporal order of the fixations), and how they move from one location to another with respect to the dynamics of the scene and the mechanics of the eyes (dynamics). State-of-the-art models focus on learning saliency maps from human data, a process that only takes into account the spatial component of the phenomenon and ignore its temporal and dynamical counterparts. In this work we focus on the evaluation methodology of models of human visual attention. We underline the limits of the current metrics for saliency prediction and scanpath similarity, and we introduce a statistical measure for the evaluation of the dynamics of the simulated eye movements. While deep learning models achieve astonishing performance in saliency prediction, our analysis shows their limitations in capturing the dynamics of the process. We find that unsupervised gravitational models, despite of their simplicity, outperform all competitors. Finally, exploiting a crowd-sourcing platform, we present a study aimed at evaluating how strongly the scanpaths generated with the unsupervised gravitational models appear plausible to naive and expert human observers.
Cloud to Street；Rutgers University（罗格斯大学）；European Space Agency（欧洲空间局）
To address the mounting destruction caused by floods in climate-vulnerable regions, we propose Street to Cloud, a machine learning pipeline for incorporating crowdsourced ground truth data into the segmentation of satellite imagery of floods. We propose this approach as a solution to the labor-intensive task of generating high-quality, hand-labeled training data, and demonstrate successes and failures of different plausible crowdsourcing approaches in our model. Street to Cloud leverages community reporting and machine learning to generate novel, near-real time insights into the extent of floo…
Inst. of Comput. Tech., CAS（中国科学院计算技术研究所）；Inst. of Info. Engin., CAS（中国科学院信息工程研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Hong Kong University of Science and Technology（香港科技大学）
The problem of estimating subjective visual properties (SVP) of images (e.g., Shoes A is more comfortable than B) is gaining rising attention. Due to its highly subjective nature, different annotators often exhibit different interpretations of scales when adopting absolute value tests. Therefore, recent investigations turn to collect pairwise comparisons via crowdsourcing platforms. However, crowdsourcing data usually contains outliers. For this purpose, it is desired to develop a robust model for learning SVP from crowdsourced noisy annotations. In this paper, we construct a deep SVP predicti…
University of Pittsburgh（匹兹堡大学）；Carnegie Mellon University（卡内基梅隆大学）；Stanford University（斯坦福大学）；University of Texas at Austin（德克萨斯大学奥斯汀分校）
Computer vision systems require large amounts of manually annotated data to properly learn challenging visual concepts. Crowdsourcing platforms offer an inexpensive method to capture human knowledge and understanding, for a vast number of visual perception tasks. In this survey, we describe the types of annotations computer vision researchers have collected using crowdsourcing, and how they have ensured that this data is of high quality while annotation effort is minimized. We begin by discussing data collection on both classic (e.g., object recognition) and recent (e.g., visual story-telling)…
In this paper, we present BubbleView, an alternative methodology for eye tracking using discrete mouse clicks to measure which information people consciously choose to examine. BubbleView is a mouse-contingent, moving-window interface in which participants are presented with a series of blurred images and click to reveal "bubbles" - small, circular areas of the image at original resolution, similar to having a confined area of focus like the eye fovea. Across 10 experiments with 28 different parameter combinations, we evaluated BubbleView on a variety of image types: information visualizations…
Scaling Audio-Visual Quality Assessment Dataset via Crowdsourcing
College of Computing and Data Science, Nanyang Technological University（计算与数据科学学院，南洋理工大学）；School of Information Science and Engineering, Shandong Normal University（信息科学与工程学院，山东师范大学）；Institute of Information Science, Beijing Jiao Tong University（信息科学研究院，北京交通大学）；YouTube Media Algorithms team, Google Inc.（YouTube媒体算法团队，谷歌公司）
Visual question answering (VQA) systems are emerging from a desire to empower users to ask any natural language question about visual content and receive a valid answer in response. However, close examination of the VQA problem reveals an unavoidable, entangled problem that multiple humans may or may not always agree on a single answer to a visual question. We train a model to automatically predict from a visual question whether a crowd would agree on a single answer. We then propose how to exploit this system in a novel application to efficiently allocate human effort to collect answers to vi…
People's visual experiences of the world are easy to carve up and examine along natural language boundaries, e.g., by category labels, attribute labels, etc. However, it is more difficult to elicit detailed visuospatial information about what a person attends to, e.g., the specific shape of a tree. Paying attention to the shapes of things not only feeds into well defined tasks like visual category learning, but it is also what enables us to differentiate similarly named objects and to take on creative visual pursuits, like poetically describing the shape of a thing, or finding shapes in the cl…
Michael Stifel Center Jena（迈克尔·斯蒂费尔中心）；Friedrich-Schiller-University Jena（耶拿弗里德里希·席勒大学）
Fine-grained visual categorization is a classification task for distinguishing categories with high intra-class and small inter-class variance. While global approaches aim at using the whole image for performing the classification, part-based solutions gather additional local information in terms of attentions or parts. We propose a novel classification-specific part estimation that uses an initial prediction as well as back-propagation of feature importance via gradient computations in order to estimate relevant image regions. The subsequently detected parts are then not only selected by a-po…
The ability of to explain neural network decisions goes hand in hand with their safe deployment. Several methods have been proposed to highlight features important for a given network decision. However, there is no consensus on how to measure effectiveness of these methods. We propose a new procedure for evaluating explanations. We use it to investigate visual explanations extracted from a range of possible sources in a neural network. We quantify the benefit of combining these sources and challenge a recent appeal for taking bias parameters into account. We support our conclusions with a gene…
Interpretation and explanation of deep models is critical towards wide adoption of systems that rely on them. In this paper, we propose a novel scheme for both interpretation as well as explanation in which, given a pretrained model, we automatically identify internal features relevant for the set of classes considered by the model, without relying on additional annotations. We interpret the model through average visualizations of this reduced set of features. Then, at test time, we explain the network prediction by accompanying the predicted class label with supporting visualizations derived …
Institute of Image Communication and Network Engineering, Shanghai JiaoTong University（上海交通大学图像通信与网络工程研究所）
Traditional adversarial examples are typically generated by adding perturbation noise to the input image within a small matrix norm. In practice, un-restricted adversarial attack has raised great concern and presented a new threat to the AI safety. In this paper, we propose a wavelet-VAE structure to reconstruct an input image and generate adversarial examples by modifying the latent code. Different from perturbation-based attack, the modifications of the proposed method are not limited but imperceptible to human eyes. Experiments show that our method can generate high quality adversarial exam…
University of Toronto（多伦多大学）；Vector Institute（矢量研究院）；Google Brain（谷歌大脑）
We present a new technique for learning visual-semantic embeddings for cross-modal retrieval. Inspired by hard negative mining, the use of hard negatives in structured prediction, and ranking loss functions, we introduce a simple change to common loss functions used for multi-modal embeddings. That, combined with fine-tuning and use of augmented data, yields significant gains in retrieval performance. We showcase our approach, VSE++, on MS-COCO and Flickr30K datasets, using ablation studies and comparisons with existing methods. On MS-COCO our approach outperforms state-of-the-art methods by 8…
Oxford NIHR Biomedical Research Centre (BRC)（牛津国家健康研究所生物医学研究中心）；John Radcliffe Hospital（约翰·拉德克利夫医院）
While challenging, the dense segmentation of histology images is a necessary first step to assess changes in tissue architecture and cellular morphology. Although specific convolutional neural network architectures have been applied with great success to the problem, few effectively incorporate visual context information from multiple scales. With this paper, we present a systematic comparison of different architectures to assess how including multi-scale information affects segmentation performance. A publicly available breast cancer and a locally collected prostate cancer datasets are being …
Convolutional Neural Networks (CNNs) for visual tasks are believed to learn both the low-level textures and high-level object attributes, throughout the network depth. This paper further investigates the `texture bias' in CNNs. To this end, we regenerate multiple instances of training examples from each original image, through a process we call `repainting'. The repainted examples preserve the shape and structure of the regions and objects within the scenes, but diversify their texture and color. Our method can regenerate a same image at different daylight, season, or weather conditions, can h…
We introduce a state-of-the-art audio-visual on-screen sound separation system which is capable of learning to separate sounds and associate them with on-screen objects by looking at in-the-wild videos. We identify limitations of previous work on audio-visual on-screen sound separation, including the simplicity and coarse resolution of spatio-temporal attention, and poor convergence of the audio separation model. Our proposed model addresses these issues using cross-modal and self-attention modules that capture audio-visual dependencies at a finer resolution over time, and by unsupervised pre-…
Weakly supervised visual grounding aims to predict the region in an image that corresponds to a specific linguistic query, where the mapping between the target object and query is unknown in the training stage. The state-of-the-art method uses a vision language pre-training model to acquire heatmaps from Grad-CAM, which matches every query word with an image region, and uses the combined heatmap to rank the region proposals. In this paper, we propose two simple but efficient methods for improving this approach. First, we propose a target-aware cropping approach to encourage the model to learn …
University of California San Diego（加州大学圣地亚哥分校）；Korea Electronics Technology Institute（韩国电子技术研究院）；Cornell NYC Tech（康奈尔纽约科技学院）
Despite recent advances in video segmentation, many opportunities remain to improve it using a variety of low and mid-level visual cues. We propose improvements to the leading streaming graph-based hierarchical video segmentation (streamGBH) method based on early and mid level visual processing. The extensive experimental analysis of our approach validates the improvement of hierarchical supervoxel representation by incorporating motion and color with effective filtering. We also pose and illuminate some open questions towards intermediate level video analysis as further extension to streamGBH…
Humans learn language by listening, speaking, writing, reading, and also, via interaction with the multimodal real world. Existing language pre-training frameworks show the effectiveness of text-only self-supervision while we explore the idea of a visually-supervised language model in this paper. We find that the main reason hindering this exploration is the large divergence in magnitude and distributions between the visually-grounded language datasets and pure-language corpora. Therefore, we develop a technique named "vokenization" that extrapolates multimodal alignments to language-only data…
Pose estimation and map building are central ingredients of autonomous robots and typically rely on the registration of sensor data. In this paper, we investigate a new metric for registering images that builds upon on the idea of the photometric error. Our approach combines a gradient orientation-based metric with a magnitude-dependent scaling term. We integrate both into stereo estimation as well as visual odometry systems and show clear benefits for typical disparity and direct image registration tasks when using our proposed metric. Our experimental evaluation indicats that our metric lead…
The University of Adelaide（阿德莱德大学）；The University of Hong Kong（香港大学）；Northwest A&F University（西北农林科技大学）；CSIRO Data61（联邦科学与工业研究组织数据61研究所）；Zhejiang University（浙江大学）
Monocular visual odometry (VO) is an important task in robotics and computer vision. Thus far, how to build accurate and robust monocular VO systems that can work well in diverse scenarios remains largely unsolved. In this paper, we propose a framework to exploit monocular depth estimation for improving VO. The core of our framework is a monocular depth estimation module with a strong generalization capability for diverse scenes. It consists of two separate working modes to assist the localization and mapping. With a single monocular image input, the depth estimation module predicts a relative…
Our brains combine vision and hearing to create a more elaborate interpretation of the world. When the visual input is insufficient, a rich panoply of sounds can be used to describe our surroundings. Since more than 1,000 hours of videos are uploaded to the internet everyday, it is arduous, if not impossible, to manually annotate these videos. Therefore, incorporating audio along with visual data without annotations is crucial for leveraging this explosion of data for recognizing and understanding objects and scenes. Owens,et.al suggest that a rich representation of the physical world can be l…
Zero-shot learning (ZSL) highly depends on a good semantic embedding to connect the seen and unseen classes. Recently, distributed word embeddings (DWE) pre-trained from large text corpus have become a popular choice to draw such a connection. Compared with human defined attributes, DWEs are more scalable and easier to obtain. However, they are designed to reflect semantic similarity rather than visual similarity and thus using them in ZSL often leads to inferior performance. To overcome this visual-semantic discrepancy, this work proposes an objective function to re-align the distributed word…
Over the last decade, Computer Vision, the branch of Artificial Intelligence aimed at understanding the visual world, has evolved from simply recognizing objects in images to describing pictures, answering questions about images, aiding robots maneuver around physical spaces and even generating novel visual content. As these tasks and applications have modernized, so too has the reliance on more data, either for model training or for evaluation. In this chapter, we demonstrate that novel interaction strategies can enable new forms of data collection and evaluation for Computer Vision. First, w…
Nanyang Technological University（南洋理工大学）；Rice University（莱斯大学）；Alibaba AI Labs（阿里巴巴人工智能实验室）
Textual-visual cross-modal retrieval has been a hot research topic in both computer vision and natural language processing communities. Learning appropriate representations for multi-modal data is crucial for the cross-modal retrieval performance. Unlike existing image-text retrieval approaches that embed image-text pairs as single feature vectors in a common representational space, we propose to incorporate generative processes into the cross-modal feature embedding, through which we are able to learn not only the global abstract features but also the local grounded features. Extensive experi…
The Chinese University of Hong Kong（香港中文大学）；SenseTime Research（商汤科技研究院）；Xi’an Jiaotong University（西安交通大学）
Person re-identification is an important task that requires learning discriminative visual features for distinguishing different person identities. Diverse auxiliary information has been utilized to improve the visual feature learning. In this paper, we propose to exploit natural language description as additional training supervisions for effective visual features. Compared with other auxiliary information, language can describe a specific person from more compact and semantic visual aspects, thus is complementary to the pixel-level image data. Our method not only learns better global visual …
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；CAS Center for Excellence in Brain Science and Intelligence Technology（中国科学院脑科学与智能技术卓越创新中心）；The Chinese University of Hong Kong（香港中文大学）
Visual grounding is a task to locate the target indicated by a natural language expression. Existing methods extend the generic object detection framework to this problem. They base the visual grounding on the features from pre-generated proposals or anchors, and fuse these features with the text embeddings to locate the target mentioned by the text. However, modeling the visual features from these predefined locations may fail to fully exploit the visual context and attribute information in the text query, which limits their performance. In this paper, we propose a transformer-based framework…
University of Rochester（罗切斯特大学）；Tencent AI Lab（腾讯人工智能实验室）
We improve one-stage visual grounding by addressing current limitations on grounding long and complex queries. Existing one-stage methods encode the entire language query as a single sentence embedding vector, e.g., taking the embedding from BERT or the hidden state from LSTM. This single vector representation is prone to overlooking the detailed descriptions in the query. To address this query modeling deficiency, we propose a recursive sub-query construction framework, which reasons between image and query for multiple rounds and reduces the referring ambiguity step by step. We show our new …
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Visual Dialog is a challenging vision-language task since the visual dialog agent needs to answer a series of questions after reasoning over both the image content and dialog history. Though existing methods try to deal with the cross-modal understanding in visual dialog, they are still not enough in ranking candidate answers based on their understanding of visual and textual contexts. In this paper, we analyze the cross-modal understanding in visual dialog based on the vision-language pre-training model VD-BERT and propose a novel approach to improve the cross-modal understanding for visual d…
We propose a self-supervised learning framework that uses unlabeled monocular video sequences to generate large-scale supervision for training a Visual Odometry (VO) frontend, a network which computes pointwise data associations across images. Our self-improving method enables a VO frontend to learn over time, unlike other VO and SLAM systems which require time-consuming hand-tuning or expensive data collection to adapt to new environments. Our proposed frontend operates on monocular images and consists of a single multi-task convolutional neural network which outputs 2D keypoints locations, k…
University of Alberta（阿尔伯塔大学）；Universiti Putra Malaysia（马来西亚博特拉大学）
Reliable feature correspondence between frames is a critical step in visual odometry (VO) and visual simultaneous localization and mapping (V-SLAM) algorithms. In comparison with existing VO and V-SLAM algorithms, semi-direct visual odometry (SVO) has two main advantages that lead to state-of-the-art frame rate camera motion estimation: direct pixel correspondence and efficient implementation of probabilistic mapping method. This paper improves the SVO mapping by initializing the mean and the variance of the depth at a feature location according to the depth prediction from a single-image dept…
OmniTaskonomy: When Does Visual Generation Improve Visual Understanding?
University of California, Berkeley（加州大学伯克利分校）；Duke University（杜克大学）；Carnegie Mellon University（卡内基梅隆大学）；University of Washington（华盛顿大学）；Elorian；Impossible Research
Attention is an important component of modern deep learning. However, less emphasis has been put on its inverse: ignoring distraction. Our daily lives require us to explicitly avoid giving attention to salient visual features that confound the task we are trying to accomplish. This visual prioritisation allows us to concentrate on important tasks while ignoring visual distractors. In this work, we introduce Neural Blindness, which gives an agent the ability to completely ignore objects or classes that are deemed distractors. More explicitly, we aim to render a neural network completely incapab…
Recent advances in deep learning have markedly improved the quality of visual-attention modelling. In this work we apply these advances to video compression. We propose a compression method that uses a saliency model to adaptively compress frame areas in accordance with their predicted saliency. We selected three state-of-the-art saliency models, adapted them for video compression and analyzed their results. The analysis includes objective evaluation of the models as well as objective and subjective evaluation of the compressed videos. Our method, which is based on the x264 video codec, can pr…
The fusion of camera sensor and inertial data is a leading method for ego-motion tracking in autonomous and smart devices. State estimation techniques that rely on non-linear filtering are a strong paradigm for solving the associated information fusion task. The de facto inference method in this space is the celebrated extended Kalman filter (EKF), which relies on first-order linearizations of both the dynamical and measurement model. This paper takes a critical look at the practical implications and limitations posed by the EKF, especially under faulty visual feature associations and the pres…
Institute for Aerospace Studies, University of Toronto（多伦多大学航空航天研究所）
In the absence of reliable and accurate GPS, visual odometry (VO) has emerged as an effective means of estimating the egomotion of robotic vehicles. Like any dead-reckoning technique, VO suffers from unbounded accumulation of drift error over time, but this accumulation can be limited by incorporating absolute orientation information from, for example, a sun sensor. In this paper, we leverage recent work on visual outdoor illumination estimation to show that estimation error in a stereo VO pipeline can be reduced by inferring the sun position from the same image stream used to compute VO, ther…
University of Arizona（亚利桑那大学）；Jet Propulsion Laboratory, California Institute of Technology（加州理工学院喷气推进实验室）
Glacial science could benefit tremendously from autonomous robots, but previous glacial robots have had perception issues in these colorless and featureless environments, specifically with visual feature extraction. This translates to failures in visual odometry and visual navigation. Glaciologists use near-infrared imagery to reveal the underlying heterogeneous spatial structure of snow and ice, and we theorize that this hidden near-infrared structure could produce more and higher quality features than available in visible light. We took a custom camera rig to Igloo Cave at Mt. St. Helens to …
Georgia Tech（佐治亚理工学院）；Facebook AI Research（Facebook人工智能研究院）
Prior work on training generative Visual Dialog models with reinforcement learning(Das et al.) has explored a Qbot-Abot image-guessing game and shown that this 'self-talk' approach can lead to improved performance at the downstream dialog-conditioned image-guessing task. However, this improvement saturates and starts degrading after a few rounds of interaction, and does not lead to a better Visual Dialog model. We find that this is due in part to repeated interactions between Qbot and Abot during self-talk, which are not informative with respect to the image. To improve this, we devise a simpl…
