ICCV 2019 Papers

We propose an end-to-end trainable Convolutional Neural Network (CNN), named GridDehazeNet, for single image dehazing. The GridDehazeNet consists of three modules: pre-processing, backbone, and post-processing. The trainable pre-processing module can generate learned inputs with better diversity and more pertinent features as compared to those derived inputs produced by hand-selected pre-processing methods. The backbone module implements a novel attention-based multi-scale estimation on a grid network, which can effectively alleviate the bottleneck issue often encountered in the conventional multi-scale approach. The post-processing module helps to reduce the artifacts in the final output. Experimental results indicate that the GridDehazeNet outperforms the state-of-the-arts on both synthetic and real-world images. The proposed hazing method does not rely on the atmosphere scattering model, and we provide an explanation as to why it is not necessarily beneficial to take advantage of the dimension reduction offered by the atmosphere scattering model for image dehazing, even if only the dehazing results on synthetic images are concerned.

Estimating the metric height of a person from monocular imagery without additional assumptions is ill-posed. Existing solutions either require manual calibration of ground plane and camera geometry, special cameras, or reference objects of known size. We focus on motion cues and exploit gravity on earth as an omnipresent reference 'object' to translate acceleration, and subsequently height, measured in image-pixels to values in meters. We require videos of motion as input, where gravity is the only external force. This limitation is different to those of existing solutions that recover a person's height and, therefore, our method opens up new application fields. We show theoretically and empirically that a simple motion trajectory analysis suffices to translate from pixel measurements to the person's metric height, reaching a MAE of up to 3.9 cm on jumping motions, and that this works without camera and ground plane calibration.

In this paper, we present a novel deep method to reconstruct a point cloud of an object from a single still image. Prior arts in the field struggle to reconstruct an accurate and scalable 3D model due to either the inefficient and expensive 3D representations, the dependency between the output and number of model parameters or the lack of a suitable computing operation. We propose to overcome these by deforming a random point cloud to the object shape through two steps: feature blending and deformation. In the first step, the global and point-specific shape features extracted from a 2D object image are blended with the encoded feature of a randomly generated point cloud, and then this mixture is sent to the deformation step to produce the final representative point set of the object. In the deformation process, we introduce a new layer termed as GraphX that considers the inter-relationship between points like common graph convolutions but operates on unordered sets. Moreover, with a simple trick, the proposed model can generate an arbitrary-sized point cloud, which is the first deep method to do so. Extensive experiments verify that we outperform existing models and halve the state-of-the-art distance score in single image 3D reconstruction.

Neuromorphic vision sensing (NVS) devices represent visual information as sequences of asynchronous discrete events (a.k.a., "spikes'") in response to changes in scene reflectance. Unlike conventional active pixel sensing (APS), NVS allows for significantly higher event sampling rates at substantially increased energy efficiency and robustness to illumination changes. However, object classification with NVS streams cannot leverage on state-of-the-art convolutional neural networks (CNNs), since NVS does not produce frame representations. To circumvent this mismatch between sensing and processing with CNNs, we propose a compact graph representation for NVS. We couple this with novel residual graph CNN architectures and show that, when trained on spatio-temporal NVS data for object classification, such residual graph CNNs preserve the spatial and temporal coherence of spike events, while requiring less computation and memory. Finally, to address the absence of large real-world NVS datasets for complex recognition tasks, we present and make available a 100k dataset of NVS recordings of the American sign language letters, acquired with an iniLabs DAVIS240c device under real-world conditions.

Most state-of-the-art action localization systems process each action proposal individually, without explicitly exploiting their relations during learning. However, the relations between proposals actually play an important role in action localization, since a meaningful action always consists of multiple proposals in a video. In this paper, we propose to exploit the proposal-proposal relations using GraphConvolutional Networks (GCNs). First, we construct an action proposal graph, where each proposal is represented as a node and their relations between two proposals as an edge. Here, we use two types of relations, one for capturing the context information for each proposal and the other one for characterizing the correlations between distinct actions. Then we apply the GCNs over the graph to model the relations among different proposals and learn powerful representations for the action classification and localization. Experimental results show that our approach significantly outperforms the state-of-the-art on THUMOS14(49.1% versus 42.8%). Moreover, augmentation experiments on ActivityNet also verify the efficacy of modeling action proposal relationships.

The fully-convolutional siamese network based on template matching has shown great potentials in visual tracking. During testing, the template is fixed with the initial target feature and the performance totally relies on the general matching ability of the siamese network. However, this manner cannot capture the temporal variations of targets or background clutter. In this work, we propose a novel gradient-guided network to exploit the discriminative information in gradients and update the template in the siamese network through feed-forward and backward operations. To be specific, the algorithm can utilize the information from the gradient to update the template in the current frame. In addition, a template generalization training method is proposed to better use gradient information and avoid overfitting. To our knowledge, this work is the first attempt to exploit the information in the gradient for template update in siamese-based trackers. Extensive experiments on recent benchmarks demonstrate that our method achieves better performance than other state-of-the-art trackers.

Automatic data abstraction is an important capability for both benchmarking machine intelligence and supporting summarization applications. In the former one asks whether a machine can `understand' enough about the meaning of input data to produce a meaningful but more compact abstraction. In the latter this capability is exploited for saving space or human time by summarizing the essence of input data. In this paper we study a general reinforcement learning based framework for learning to abstract sequential data in a goal-driven way. The ability to define different abstraction goals uniquely allows different aspects of the input data to be preserved according to the ultimate purpose of the abstraction. Our reinforcement learning objective does not require human-defined examples of ideal abstraction. Importantly our model processes the input sequence holistically without being constrained by the original input order. Our framework is also domain agnostic -- we demonstrate applications to sketch, video and text data and achieve promising results in all domains.

This paper proposes the Global-Local Temporal Representation (GLTR) to exploit the multi-scale temporal cues in video sequences for video person Re-Identification (ReID). GLTR is constructed by first modeling the short-term temporal cues among adjacent frames, then capturing the long-term relations among inconsecutive frames. Specifically, the short-term temporal cues are modeled by parallel dilated convolutions with different temporal dilation rates to represent the motion and appearance of pedestrian. The long-term relations are captured by a temporal self-attention model to alleviate the occlusions and noises in video sequences. The short and long-term temporal cues are aggregated as the final GLTR by a simple single-stream CNN. GLTR shows substantial superiority to existing features learned with body part cues or metric learning on four widely-used video ReID datasets. For instance, it achieves Rank-1 Accuracy of 87.02% on MARS dataset without re-ranking, better than current state-of-the art.

In deep convolutional neural networks (CNNs), local pooling operation is a key building block to effectively downsize feature maps for reducing computation cost as well as increasing robustness against input variation. There are several types of pooling operation, such as average/max-pooling, from which one has to be manually selected for building CNNs. The optimal pooling type would be dependent on characteristics of features in CNNs and classification tasks, making it hard to find out the proper pooling module in advance. In this paper, we propose a flexible pooling method which adaptively tunes the pooling functionality based on input features without manually fixing it beforehand. In the proposed method, the parameterized pooling form is derived from a probabilistic perspective to flexibly represent various types of pooling and then the parameters are estimated by means of global statistics in the input feature map. Thus, the proposed local pooling guided by global features effectively works in the CNNs trained in an end-to-end manner. The experimental results on image classification tasks demonstrate the effectiveness of the proposed pooling method in various deep CNNs.

Large geometry (e.g., orientation) variances are the key challenges in the scene text detection. In this work, we first conduct experiments to investigate the capacity of networks for learning geometry variances on detecting scene texts, and find that networks can handle only limited text geometry variances. Then, we put forward a novel Geometry Normalization Module (GNM) with multiple branches, each of which is composed of one Scale Normalization Unit and one Orientation Normalization Unit, to normalize each text instance to one desired canonical geometry range through at least one branch. The GNM is general and readily plugged into existing convolutional neural network based text detectors to construct end-to-end Geometry Normalization Networks (GNNets). Moreover, we propose a geometry-aware training scheme to effectively train the GNNets by sampling and augmenting text instances from a uniform geometry variance distribution. Finally, experiments on popular benchmarks of ICDAR 2015 and ICDAR 2017 MLT validate that our method outperforms all the state-of-the-art approaches remarkably by obtaining one-forward test F-scores of 88.52 and 74.54 respectively.

Representing 3D shapes is a fundamental problem in artificial intelligence, which has numerous applications within computer vision and graphics. One avenue that has recently begun to be explored is the use of latent representations of generative models. However, it remains an open problem to learn a generative model of shapes that is interpretable and easily manipulated, particularly in the absence of supervised labels. In this paper, we propose an unsupervised approach to partitioning the latent space of a variational autoencoder for 3D point clouds in a natural way, using only geometric information, that builds upon prior work utilizing generative adversarial models of point sets. Our method makes use of tools from spectral geometry to separate intrinsic and extrinsic shape information, and then considers several hierarchical disentanglement penalties for dividing the latent space in this manner. We also propose a novel disentanglement penalty that penalizes the predicted change in the latent representation of the output,with respect to the latent variables of the initial shape. We show that the resulting latent representation exhibits intuitive and interpretable behaviour, enabling tasks such as pose transfer that cannot easily be performed by models with an entangled representation.

Understanding fashion styles and trends is of great potential interest to retailers and consumers alike. The photos people upload to social media are a historical and public data source of how people dress across the world and at different times. While we now have tools to automatically recognize the clothing and style attributes of what people are wearing in these photographs, we lack the ability to analyze spatial and temporal trends in these attributes or make predictions about the future. In this paper we address this need by providing an automatic framework that analyzes large corpora of street imagery to (a) discover and forecast long-term trends of various fashion attributes as well as automatically discovered styles, and (b) identify spatio-temporally localized events that affect what people wear. We show that our framework makes long term trend forecasts that are > 20% more accurate than prior art, and identifies hundreds of socially meaningful events that impact fashion across the globe.

Multi-view action recognition targets to integrate complementary information from different views to improve classification performance. It is a challenging task due to the distinct gap between heterogeneous feature domains. Moreover, most existing methods neglect to consider the incomplete multi-view data, which limits their potential compatibility in real-world applications. In this work, we propose a Generative Multi-View Action Recognition (GMVAR) framework to address the challenges above. The adversarial generative network is leveraged to generate one view conditioning on the other view, which fully explores the latent connections in both intra-view and cross-view aspects. Our approach enhances the model robustness by employing adversarial training, and naturally handles the incomplete view case by imputing the missing data. Moreover, an effective View Correlation Discovery Network (VCDN) is proposed to further fuse the multi-view information in a higher-level label space. Extensive experiments demonstrate the effectiveness of our proposed approach by comparing with state-of-the-art algorithms.

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%.

The detection and removal of cloud in remote sensing images are essential for earth observation applications. Most previous methods consider cloud detection as a pixel-wise semantic segmentation process (cloud v.s. background), which inevitably leads to a category-ambiguity problem when dealing with semi-transparent clouds. We re-examine the cloud detection under a totally different point of view, i.e. to formulate it as a mixed energy separation process between foreground and background images, which can be equivalently implemented under an image matting paradigm with a clear physical significance. We further propose a generative adversarial framework where the training of our model neither requires any pixel-wise ground truth reference nor any additional user interactions. Our model consists of three networks, a cloud generator G, a cloud discriminator D, and a cloud matting network F, where G and D aim to generate realistic and physically meaningful cloud images by adversarial training, and F learns to predict the cloud reflectance and attenuation. Experimental results on a global set of satellite images demonstrate that our method, without ever using any pixel-wise ground truth during training, achieves comparable and even higher accuracy over other fully supervised methods, including some recent popular cloud detectors and some well-known semantic segmentation frameworks.

We present a learned image compression system based on GANs, operating at extremely low bitrates. Our proposed framework combines an encoder, decoder/generator and a multi-scale discriminator, which we train jointly for a generative learned compression objective. The model synthesizes details it cannot afford to store, obtaining visually pleasing results at bitrates where previous methods fail and show strong artifacts. Furthermore, if a semantic label map of the original image is available, our method can fully synthesize unimportant regions in the decoded image such as streets and trees from the label map, proportionally reducing the storage cost. A user study confirms that for low bitrates, our approach is preferred to state-of-the-art methods, even when they use more than double the bits.

Class imbalance is a long-standing problem relevant to a number of real-world applications of deep learning. Oversampling techniques, which are effective for handling class imbalance in classical learning systems, can not be directly applied to end-to-end deep learning systems. We propose a three-player adversarial game between a convex generator, a multi-class classifier network, and a real/fake discriminator to perform oversampling in deep learning systems. The convex generator generates new samples from the minority classes as convex combinations of existing instances, aiming to fool both the discriminator as well as the classifier into misclassifying the generated samples. Consequently, the artificial samples are generated at critical locations near the peripheries of the classes. This, in turn, adjusts the classifier induced boundaries in a way which is more likely to reduce misclassification from the minority classes. Extensive experiments on multiple class imbalanced image datasets establish the efficacy of our proposal.

This paper addresses the generation of referring expressions that not only refer to objects correctly but also let humans find them quickly. As a target becomes relatively less salient, identifying referred objects itself becomes more difficult. However, the existing studies regarded all sentences that refer to objects correctly as equally good, ignoring whether they are easily understood by humans. If the target is not salient, humans utilize relationships with the salient contexts around it to help listeners to comprehend it better. To derive this information from human annotations, our model is designed to extract information from the target and from the environment. Moreover, we regard that sentences that are easily understood are those that are comprehended correctly and quickly by humans. We optimized this by using the time required to locate the referred objects by humans and their accuracies. To evaluate our system, we created a new referring expression dataset whose images were acquired from Grand Theft Auto V (GTA V), limiting targets to persons. Experimental results show the effectiveness of our approach. Our code and dataset are available at https://github.com/mikittt/easy-to-understand-REG.

Recently there has been significant progress in image captioning with the help of deep learning. However, captions generated by current state-of-the-art models are still far from satisfactory, despite high scores in terms of conventional metrics such as BLEU and CIDEr. Human-written captions are diverse, informative and precise, but machine-generated captions seem to be simple, vague and dull. In this paper, aimed at improving diversity and descriptiveness characteristics of generated image captions, we propose a model utilizing visual paraphrases (different sentences describing the same image) in captioning datasets. We explore different strategies to select useful visual paraphrase pairs for training by designing a variety of scoring functions. Our model consists of two decoding stages, where a preliminary caption is generated in the first stage and then paraphrased into a more diverse and descriptive caption in the second stage. Extensive experiments are conducted on the benchmark MS COCO dataset, with automatic evaluation and human evaluation results verifying the effectiveness of our model.