Imitation learning by behavioral cloning is a prevalent method which has achieved some success in vision-based autonomous driving. The basic idea behind behavioral cloning is to have the neural network learn from observing a human expert’s behavior. Typically, a convolutional neural network learns to predict the steering commands from raw driver-view images by mimicking the behaviors of human drivers. However, there are other cues, e.g. gaze behavior, available from human drivers that have yet to be exploited. Previous researches have shown that novice human learners can benefit from observing experts’ gaze patterns. We present here that deep neural networks can also profit from this. We propose a method, gaze-modulated dropout, for integrating this gaze information into a deep driving network implicitly rather than as an additional input. Our experimental results demonstrate that gaze-modulated dropout enhances the generalization capability of the network to unseen scenes. Prediction error in steering commands is reduced by 23.5% compared to uniform dropout. Running closed loop in the simulator, the gaze-modulated dropout net increased the average distance travelled between infractions by 58.5%. Consistent with these results, we also found the gaze-modulated dropout net to have lower model uncertainty.
http://arxiv.org/abs/1904.08377
Signal extraction from a single-channel mixture with additional undesired signals is most commonly performed using time-frequency (TF) masks. Typically, the mask is estimated with a deep neural network (DNN) and element-wise applied to the complex mixture short-time Fourier transform (STFT) representation to perfom extraction. Ideal mask magnitudes are zero for solely undesired signals in a TF bin and infinity for total destructive interference. Usually, masks have an upper bound to provide well-defined DNN outputs at the cost of limited extraction capabilities. We propose to estimate with a DNN a complex TF filter for each mixture TF bin which maps an STFT area in the respective mixture to the desired TF bin to address destructive interference in mixture TF bins. The DNN is optimized by minimizing the error between the extracted and the groundtruth desired signal allowing to train without having to specify ground-truth TF filters but learn filters by error reduction. We compare our approach with complex and real valued TF masks by separating speech from a variety of different sound and noise classes from the Google AudioSet corpus. We also process the mixture STFT with notch filters and zero whole time-frames to demonstrate the reconstruction capabilities of our approach. The proposed method outperformed the baselines especially when notch filters and time-frame zeroing were applied.
http://arxiv.org/abs/1904.08369
We propose a novel approach for 3D shape completion by synthesizing multi-view depth maps. While previous work for shape completion relies on volumetric representations, meshes, or point clouds, we propose to use multi-view depth maps from a set of fixed viewing angles as our shape representation. This allows us to be free of the limitations of memory for volumetric representations and point clouds by casting shape completion into an image-to-image translation problem. Specifically, we render depth maps of the incomplete shape from a fixed set of viewpoints, and perform depth map completion in each view. Different from image-to-image translation network that completes each view separately, our novel network, multi-view completion net (MVCN), leverages information from all views of a 3D shape to help the completion of each single view. This enables MVCN to leverage more information from different depth views to achieve high accuracy in single depth view completion and keep the consistency among the completed depth images in different views. Benefited by the multi-view representation and the novel network structure, MVCN significantly improves the accuracy of 3D shape completion in large-scale benchmarks compared to the state of the art.
http://arxiv.org/abs/1904.08366
In this paper, we propose a novel method, aggregation cross-entropy (ACE), for sequence recognition from a brand new perspective. The ACE loss function exhibits competitive performance to CTC and the attention mechanism, with much quicker implementation (as it involves only four fundamental formulas), faster inference\back-propagation (approximately O(1) in parallel), less storage requirement (no parameter and negligible runtime memory), and convenient employment (by replacing CTC with ACE). Furthermore, the proposed ACE loss function exhibits two noteworthy properties: (1) it can be directly applied for 2D prediction by flattening the 2D prediction into 1D prediction as the input and (2) it requires only characters and their numbers in the sequence annotation for supervision, which allows it to advance beyond sequence recognition, e.g., counting problem. The code is publicly available at https://github.com/summerlvsong/Aggregation-Cross-Entropy.
http://arxiv.org/abs/1904.08364
This paper addresses the problem of learning the optimal control policy for a nonlinear stochastic dynamical system with continuous state space, continuous action space and unknown dynamics. This class of problems are typically addressed in stochastic adaptive control and reinforcement learning literature using model-based and model-free approaches respectively. Both methods rely on solving a dynamic programming problem, either directly or indirectly, for finding the optimal closed loop control policy. The inherent curse of dimensionality' associated with dynamic programming method makes these approaches also computationally difficult.
This paper proposes a novel decoupled data-based control (D2C) algorithm that addresses this problem using a decoupled, open loop - closed loop’, approach. First, an open-loop deterministic trajectory optimization problem is solved using a black-box simulation model of the dynamical system. Then, a closed loop control is developed around this open loop trajectory by linearization of the dynamics about this nominal trajectory. By virtue of linearization, a linear quadratic regulator based algorithm can be used for this closed loop control. We show that the performance of D2C algorithm is approximately optimal. Moreover, simulation performance suggests significant reduction in training time compared to other state of the art algorithms.
http://arxiv.org/abs/1904.08361
Existing objective evaluation metrics for voice conversion (VC) are not always correlated well with human perception. Therefore, training VC models with such criteria may not effectively improve naturalness and similarity of converted speech. In this paper, we propose deep learning based assessment models to predict human ratings of converted speech. We adopt the convolutional and recurrent neural network models to build a mean opinion score (MOS) predictor, termed MOSNet. The proposed models are tested on the large-scale listening test results of the Voice Conversion Challenge (VCC) 2018. Experimental results show that the predicted scores of the proposed MOSNet are highly correlated with human MOS ratings at the system level while fairly correlated with human MOS ratings at the utterance level. Meanwhile, we have modified MOSNet to predict the similarity scores, and the preliminary results show that the predicted scores are also fairly correlated with human ratings. These results confirm that the proposed models could be used as a computational evaluator to measure the MOS of VC systems to reduce the need for expensive human rating.
http://arxiv.org/abs/1904.08352
This technical report describes a practical field test on word-image classification in a very large collection of more than 300 diverse handwritten historical manuscripts, with 1.6 million unique labeled images and more than 11 million images used in testing. Results indicate that several deep-learning tests completely failed (mean accuracy 83%). In the tests with more than 1000 output units (lexical words) in one-hot encoding for classification, performance steeply drops to almost zero percent accuracy, even with a modest size of the pre-final (i.e., penultimate) layer (150 units). A traditional feature method (BOVW) displays a consistent performance over numbers of classes and numbers of training examples (mean accuracy 87%). Additional tests using nearest mean on the output of the pre-final layer of an Inception V3 network, for each book, only yielded mediocre results (mean accuracy 49\%), but was not sensitive to high numbers of classes. Notably, this experiment was only possible on the basis of labels that were harvested on the basis of a traditional method which already works starting from a single labeled image per class. It is expected that the performance of the failed deep learning tests can be repaired, but only on the basis of human handcrafting (sic) of network architecture and hyperparameters. When the failed problematic books are not considered, end-to-end CNN training yields about 95% accuracy. This average is dominated by a large subset of Chinese characters, performances for other script styles being lower.
http://arxiv.org/abs/1904.08421
Visual Question Answering (VQA) faces two major challenges: how to better fuse the visual and textual modalities and how to make the VQA model have the reasoning ability to answer more complex questions. In this paper, we address both challenges by proposing the novel Question Guided Modular Routing Networks (QGMRN). QGMRN can fuse the visual and textual modalities in multiple semantic levels which makes the fusion occur in a fine-grained way, it also can learn to reason by routing between the generic modules without additional supervision information or prior knowledge. The proposed QGMRN consists of three sub-networks: visual network, textual network and routing network. The routing network selectively executes each module in the visual network according to the pathway activated by the question features generated by the textual network. Experiments on the CLEVR dataset show that our model can outperform the state-of-the-art. Models and Codes will be released.
http://arxiv.org/abs/1904.08324
Opinions and attitudes towards controversial social and political issues are hardly ever based on evidence alone. Moral values play a fundamental role in the decision-making process of how we perceive and interpret information. The Moral Foundations Dictionary (MFD) was developed to operationalize moral values in text. In this study, we present MoralStrength, a lexicon of approximately 1,000 lemmas, obtained as an extension of the Moral Foundations Dictionary, based on WordNet synsets. Moreover, for each lemma it provides with a crowdsourced numeric assessment of Moral Valence, indicating the strength with which a lemma is expressing the specific value. We evaluated the predictive potentials of this moral lexicon, defining three utilization approaches of increasing complexity, ranging from statistical properties of the lexicon to a deep learning approach of word embeddings based on semantic similarity. Logistic regression models trained on the features extracted from MoralStrength, significantly outperformed the current state-of-the-art, reaching an F1-score of 87.6% over the previous 62.4% (p-value<0.01). Such findings pave the way for further research, allowing for an in-depth understanding of moral narratives in text for a wide range of social issues.
http://arxiv.org/abs/1904.08314
Conventional automatic speech recognition (ASR) systems trained from frame-level alignments can easily leverage posterior fusion to improve ASR accuracy and build a better single model with knowledge distillation. End-to-end ASR systems trained using the Connectionist Temporal Classification (CTC) loss do not require frame-level alignment and hence simplify model training. However, sparse and arbitrary posterior spike timings from CTC models pose a new set of challenges in posterior fusion from multiple models and knowledge distillation between CTC models. We propose a method to train a CTC model so that its spike timings are guided to align with those of a pre-trained guiding CTC model. As a result, all models that share the same guiding model have aligned spike timings. We show the advantage of our method in various scenarios including posterior fusion of CTC models and knowledge distillation between CTC models with different architectures. With the 300-hour Switchboard training data, the single word CTC model distilled from multiple models improved the word error rates to 13.7%/23.1% from 14.9%/24.1% on the Hub5 2000 Switchboard/CallHome test sets without using any data augmentation, language model, or complex decoder.
http://arxiv.org/abs/1904.08311
We study the multi-round response generation in visual dialog, where a response is generated according to a visually grounded conversational history. Given a triplet: an image, Q&A history, and current question, all the prevailing methods follow a codec (i.e., encoder-decoder) fashion in a supervised learning paradigm: a multimodal encoder encodes the triplet into a feature vector, which is then fed into the decoder for the current answer generation, supervised by the ground-truth. However, this conventional supervised learning does NOT take into account the impact of imperfect history, violating the conversational nature of visual dialog and thus making the codec more inclined to learn history bias but not contextual reasoning. To this end, inspired by the actor-critic policy gradient in reinforcement learning, we propose a novel training paradigm called History Advantage Sequence Training (HAST). Specifically, we intentionally impose wrong answers in the history, obtaining an adverse critic, and see how the historic error impacts the codec’s future behavior by History Advantage-a quantity obtained by subtracting the adverse critic from the gold reward of ground-truth history. Moreover, to make the codec more sensitive to the history, we propose a novel attention network called History-Aware Co-Attention Network (HACAN) which can be effectively trained by using HAST. Experimental results on three benchmarks: VisDial v0.9&v1.0 and GuessWhat?!, show that the proposed HAST strategy consistently outperforms the state-of-the-art supervised counterparts.
http://arxiv.org/abs/1902.09326
Abstract Meaning Representation (AMR) represents sentences as directed, acyclic and rooted graphs, aiming at capturing their meaning in a machine readable format. AMR parsing converts natural language sentences into such graphs. However, evaluating a parser on new data by means of comparison to manually created AMR graphs is very costly. Also, we would like to be able to detect parses of questionable quality, or preferring results of alternative systems by selecting the ones for which we can assess good quality. We propose AMR accuracy prediction as the task of predicting several metrics of correctness for an automatically generated AMR parse - in absence of the corresponding gold parse. We develop a neural end-to-end multi-output regression model and perform three case studies: firstly, we evaluate the model’s capacity of predicting AMR parse accuracies and test whether it can reliably assign high scores to gold parses. Secondly, we perform parse selection based on predicted parse accuracies of candidate parses from alternative systems, with the aim of improving overall results. Finally, we predict system ranks for submissions from two AMR shared tasks on the basis of their predicted parse accuracy averages. All experiments are carried out across two different domains and show that our method is effective.
http://arxiv.org/abs/1904.08301
Event cameras are novel sensors that report brightness changes in the form of asynchronous “events” instead of intensity frames. They have significant advantages over conventional cameras: high temporal resolution, high dynamic range, and no motion blur. Since the output of event cameras is fundamentally different from conventional cameras, it is commonly accepted that they require the development of specialized algorithms to accommodate the particular nature of events. In this work, we take a different view and propose to apply existing, mature computer vision techniques to videos reconstructed from event data. We propose a novel recurrent network to reconstruct videos from a stream of events, and train it on a large amount of simulated event data. Our experiments show that our approach surpasses state-of-the-art reconstruction methods by a large margin (> 20%) in terms of image quality. We further apply off-the-shelf computer vision algorithms to videos reconstructed from event data on tasks such as object classification and visual-inertial odometry, and show that this strategy consistently outperforms algorithms that were specifically designed for event data. We believe that our approach opens the door to bringing the outstanding properties of event cameras to an entirely new range of tasks. A video of the experiments is available at https://youtu.be/IdYrC4cUO0I
http://arxiv.org/abs/1904.08298
This article describes Amobee’s participation in “HatEval: Multilingual detection of hate speech against immigrants and women in Twitter” (task 5) and “OffensEval: Identifying and Categorizing Offensive Language in Social Media” (task 6). The goal of task 5 was to detect hate speech targeted to women and immigrants. The goal of task 6 was to identify and categorized offensive language in social media, and identify offense target. We present a novel type of convolutional neural network called “Multiple Choice CNN” (MC-CNN) that we used over our newly developed contextual embedding, Rozental et al. (2019). For both tasks we used this architecture and achieved 4th place out of 69 participants with an F1 score of 0.53 in task 5, in task 6 achieved 2nd place (out of 75) in Sub-task B - automatic categorization of offense types (our model reached places 18/2/7 out of 103/75/65 for sub-tasks A, B and C respectively in task 6).
http://arxiv.org/abs/1904.08292
We address the problem of defining graph transformations by the simultaneous application of direct transformations even when these cannot be applied independently of each other. An algebraic approach is adopted, with production rules of the form $L\xleftarrow{l}K \xleftarrow{i} I \xrightarrow{r} R$, called weak spans. A parallel coherent transformation is introduced and shown to be a conservative extension of the interleaving semantics of parallel independent direct transformations. A categorical construction of finitely attributed structures is proposed, in which parallel coherent transformations can be built in a natural way. These notions are introduced and illustrated on detailed examples.
http://arxiv.org/abs/1904.08850
Deep computer vision systems being vulnerable to imperceptible and carefully crafted noise have raised questions regarding the robustness of their decisions. We take a step back and approach this problem from an orthogonal direction. We propose to enable black-box neural networks to justify their reasoning both for clean and for adversarial examples by leveraging attributes, i.e. visually discriminative properties of objects. We rank attributes based on their class relevance, i.e. how the classification decision changes when the input is visually slightly perturbed, as well as image relevance, i.e. how well the attributes can be localized on both clean and perturbed images. We present comprehensive experiments for attribute prediction, adversarial example generation, adversarially robust learning, and their qualitative and quantitative analysis using predicted attributes on three benchmark datasets.
http://arxiv.org/abs/1904.08279
Data-driven generative 3D face models are used to compactly encode facial shape data into meaningful parametric representations. A desirable property of these models is their ability to effectively decouple natural sources of variation, in particular identity and expression. While factorized representations have been proposed for that purpose, they are still limited in the variability they can capture and may present modeling artifacts when applied to tasks such as expression transfer. In this work, we explore a new direction with Generative Adversarial Networks and show that they contribute to better face modeling performances, especially in decoupling natural factors, while also achieving more diverse samples. To train the model we introduce a novel architecture that combines a 3D generator with a 2D discriminator that leverages conventional CNNs, where the two components are bridged by a geometry mapping layer. We further present a training scheme, based on auxiliary classifiers, to explicitly disentangle identity and expression attributes. Through quantitative and qualitative results on standard face datasets, we illustrate the benefits of our model and demonstrate that it outperforms competing state of the art methods in terms of decoupling and diversity.
http://arxiv.org/abs/1902.03619
Hyperspectral image (HSI) consists of hundreds of continuous narrow bands with high spectral correlation, which would lead to the so-called Hughes phenomenon and the high computational cost in processing. Band selection has been proven effective in avoiding such problems by removing the redundant bands. However, many of existing band selection methods separately estimate the significance for every single band and cannot fully consider the nonlinear and global interaction between spectral bands. In this paper, by assuming that a complete HSI can be reconstructed from its few informative bands, we propose a general band selection framework, Band Selection Network (termed as BS-Net). The framework consists of a band attention module (BAM), which aims to explicitly model the nonlinear inter-dependencies between spectral bands, and a reconstruction network (RecNet), which is used to restore the original HSI cube from the learned informative bands, resulting in a flexible architecture. The resulting framework is end-to-end trainable, making it easier to train from scratch and to combine with existing networks. We implement two BS-Nets respectively using fully connected networks (BS-Net-FC) and convolutional neural networks (BS-Net-Conv), and compare the results with many existing band selection approaches for three real hyperspectral images, demonstrating that the proposed BS-Nets can accurately select informative band subset with less redundancy and achieve significantly better classification performance with an acceptable time cost.
http://arxiv.org/abs/1904.08269
In this paper, we present a novel unsupervised video summarization model that requires no manual annotation. The proposed model termed Cycle-SUM adopts a new cycle-consistent adversarial LSTM architecture that can effectively maximize the information preserving and compactness of the summary video. It consists of a frame selector and a cycle-consistent learning based evaluator. The selector is a bi-direction LSTM network that learns video representations that embed the long-range relationships among video frames. The evaluator defines a learnable information preserving metric between original video and summary video and “supervises” the selector to identify the most informative frames to form the summary video. In particular, the evaluator is composed of two generative adversarial networks (GANs), in which the forward GAN is learned to reconstruct original video from summary video while the backward GAN learns to invert the processing. The consistency between the output of such cycle learning is adopted as the information preserving metric for video summarization. We demonstrate the close relation between mutual information maximization and such cycle learning procedure. Experiments on two video summarization benchmark datasets validate the state-of-the-art performance and superiority of the Cycle-SUM model over previous baselines.
http://arxiv.org/abs/1904.08265
Prostate cancer is the most common malignant tumors in men but prostate Magnetic Resonance Imaging (MRI) analysis remains challenging. Besides whole prostate gland segmentation, the capability to differentiate between the blurry boundary of the Central Gland (CG) and Peripheral Zone (PZ) can lead to differential diagnosis, since tumor’s frequency and severity differ in these regions. To tackle the prostate zonal segmentation task, we propose a novel Convolutional Neural Network (CNN), called USE-Net, which incorporates Squeeze-and-Excitation (SE) blocks into U-Net. Especially, the SE blocks are added after every Encoder (Enc USE-Net) or Encoder-Decoder block (Enc-Dec USE-Net). This study evaluates the generalization ability of CNN-based architectures on three T2-weighted MRI datasets, each one consisting of a different number of patients and heterogeneous image characteristics, collected by different institutions. The following mixed scheme is used for training/testing: (i) training on either each individual dataset or multiple prostate MRI datasets and (ii) testing on all three datasets with all possible training/testing combinations. USE-Net is compared against three state-of-the-art CNN-based architectures (i.e., U-Net, pix2pix, and Mixed-Scale Dense Network), along with a semi-automatic continuous max-flow model. The results show that training on the union of the datasets generally outperforms training on each dataset separately, allowing for both intra-/cross-dataset generalization. Enc USE-Net shows good overall generalization under any training condition, while Enc-Dec USE-Net remarkably outperforms the other methods when trained on all datasets. These findings reveal that the SE blocks’ adaptive feature recalibration provides excellent cross-dataset generalization when testing is performed on samples of the datasets used during training.
http://arxiv.org/abs/1904.08254
Event cameras are vision sensors that record asynchronous streams of per-pixel brightness changes, referred to as “events”. They have appealing advantages over frame-based cameras for computer vision, including high temporal resolution, high dynamic range, and no motion blur. Due to the sparse, non-uniform spatiotemporal layout of the event signal, pattern recognition algorithms typically aggregate events into a grid-based representation and subsequently process it by a standard vision pipeline, e.g., Convolutional Neural Network (CNN). In this work, we introduce a general framework to convert event streams into grid-based representations through a sequence of differentiable operations. Our framework comes with two main ad-vantages: (i) allows learning the input event representation together with the task dedicated network in an end to end manner, and (ii) lays out a taxonomy that unifies the majority of extant event representations in the literature and identifies novel ones. Empirically, we show that our approach to learning the event representation end-to-end yields an improvement of approximately 12% on optical flow estimation and object recognition over state-of-the-art methods.
http://arxiv.org/abs/1904.08245
We present a novel deep convolutional network pipeline, LO-Net, for real-time lidar odometry estimation. Unlike most existing lidar odometry (LO) estimations that go through individually designed feature selection, feature matching, and pose estimation pipeline, LO-Net can be trained in an end-to-end manner. With a new mask-weighted geometric constraint loss, LO-Net can effectively learn feature representation for LO estimation, and can implicitly exploit the sequential dependencies and dynamics in the data. We also design a scan-to-map module, which uses the geometric and semantic information learned in LO-Net, to improve the estimation accuracy. Experiments on benchmark datasets demonstrate that LO-Net outperforms existing learning based approaches and has similar accuracy with the state-of-the-art geometry-based approach, LOAM.
http://arxiv.org/abs/1904.08242
Face recognition has achieved unprecedented results, surpassing human capabilities in certain scenarios. However, these automatic solutions are not ready for production because they can be easily fooled by simple identity impersonation attacks. And although much effort has been devoted to develop face anti-spoofing models, their generalization capacity still remains a challenge in real scenarios. In this paper, we introduce a novel approach that reformulates the Generalized Presentation Attack Detection (GPAD) problem from an anomaly detection perspective. Technically, a deep metric learning model is proposed, where a triplet focal loss is used as a regularization for a novel loss coined “metric-softmax”, which is in charge of guiding the learning process towards more discriminative feature representations in an embedding space. Finally, we demonstrate the benefits of our deep anomaly detection architecture, by introducing a few-shot a posteriori probability estimation that does not need any classifier to be trained on the learned features. We conduct extensive experiments using the GRAD-GPAD framework that provides the largest aggregated dataset for face GPAD. Results confirm that our approach is able to outperform all the state-of-the-art methods by a considerable margin.
http://arxiv.org/abs/1904.08241
Despite the great progress in quadrupedal robotics during the last decade, selecting good contacts (footholds) in highly uneven and cluttered environments still remains an open challenge. This paper builds upon a state-of-the-art approach, already successfully used for humanoid robots, and applies it to our robotic platform; the quadruped robot ANY-mal. The proposed algorithm decouples the problem into two subprob-lems: first a guide trajectory for the robot is generated, then contacts are created along this trajectory. Both subproblems rely on approximations and heuristics that need to be tuned. The main contribution of this work is to explain how this algorithm has been retuned to work with ANY-mal and to show the relevance of the approach with a variety of tests in realistic dynamic simulations.
http://arxiv.org/abs/1904.08238
In order to create an image segmentation method robust to lighting changes, two novel homogeneity criteria of an image region were studied. Both were defined using the Logarithmic Image Processing (LIP) framework whose laws model lighting changes. The first criterion estimates the LIP-additive homogeneity and is based on the LIP-additive law. It is theoretically insensitive to lighting changes caused by variations of the camera exposure-time or source intensity. The second, the LIP-multiplicative homogeneity criterion, is based on the LIP-multiplicative law and is insensitive to changes due to variations of the object thickness or opacity. Each criterion is then applied in Revol and Jourlin’s (1997) region growing method which is based on the homogeneity of an image region. The region growing method becomes therefore robust to the lighting changes specific to each criterion. Experiments on simulated and on real images presenting lighting variations prove the robustness of the criteria to those variations. Compared to a state-of the art method based on the image component-tree, ours is more robust. These results open the way to numerous applications where the lighting is uncontrolled or partially controlled.
http://arxiv.org/abs/1904.12597
In order to estimate a registration model of eye fundus images made of an affinity and two radial distortions, we introduce an estimation criterion based on an error between the vessels. In [1], we estimated this model by minimising the error between characteristics points. In this paper, the detected vessels are selected using the circle and ellipse equations of the overlap area boundaries deduced from our model. Our method successfully registers 96 % of the 271 pairs in a Public Health dataset acquired mostly with different cameras. This is better than our previous method [1] and better than three other state-of-the-art methods. On a publicly available dataset, ours still better register the images than the reference method.
http://arxiv.org/abs/1904.12733
We present a self-supervised approach to estimate flow in camera image and top-view grid map sequences using fully convolutional neural networks in the domain of automated driving. We extend existing approaches for self-supervised optical flow estimation by adding a regularizer expressing motion consistency assuming a static environment. However, as this assumption is violated for other moving traffic participants we also estimate a mask to scale this regularization. Adding a regularization towards motion consistency improves convergence and flow estimation accuracy. Furthermore, we scale the errors due to spatial flow inconsistency by a mask that we derive from the motion mask. This improves accuracy in regions where the flow drastically changes due to a better separation between static and dynamic environment. We apply our approach to optical flow estimation from camera image sequences, validate on odometry estimation and suggest a method to iteratively increase optical flow estimation accuracy using the generated motion masks. Finally, we provide quantitative and qualitative results based on the KITTI odometry and tracking benchmark for scene flow estimation based on grid map sequences. We show that we can improve accuracy and convergence when applying motion and spatial consistency regularization.
http://arxiv.org/abs/1904.12599
Large scale datasets created from user labels or openly available data have become crucial to provide training data for large scale learning algorithms. While these datasets are easier to acquire, the data are frequently noisy and unreliable, which is motivating research on weakly supervised learning techniques. In this paper we propose an iterative learning method that extracts the useful information from a large scale change detection dataset generated from open vector data to train a fully convolutional network which surpasses the performance obtained by naive supervised learning. We also propose the guided anisotropic diffusion algorithm, which improves semantic segmentation results using the input images as guides to perform edge preserving filtering, and is used in conjunction with the iterative training method to improve results.
http://arxiv.org/abs/1904.08208
The development of high-dimensional generative models has recently gained a great surge of interest with the introduction of variational auto-encoders and generative adversarial neural networks. Different variants have been proposed where the underlying latent space is structured, for example, based on attributes describing the data to generate. We focus on a particular problem where one aims at generating samples corresponding to a number of objects under various views. We assume that the distribution of the data is driven by two independent latent factors: the content, which represents the intrinsic features of an object, and the view, which stands for the settings of a particular observation of that object. Therefore, we propose a generative model and a conditional variant built on such a disentangled latent space. This approach allows us to generate realistic samples corresponding to various objects in a high variety of views. Unlike many multi-view approaches, our model doesn’t need any supervision on the views but only on the content. Compared to other conditional generation approaches that are mostly based on binary or categorical attributes, we make no such assumption about the factors of variations. Our model can be used on problems with a huge, potentially infinite, number of categories. We experiment it on four image datasets on which we demonstrate the effectiveness of the model and its ability to generalize.
http://arxiv.org/abs/1711.00305
Advances in variational inference enable parameterisation of probabilistic models by deep neural networks. This combines the statistical transparency of the probabilistic modelling framework with the representational power of deep learning. Yet, it seems difficult to effectively estimate such models in the context of language modelling. Even models based on rather simple generative stories struggle to make use of additional structure due to a problem known as posterior collapse. We concentrate on one such model, namely, a variational auto-encoder, which we argue is an important building block in hierarchical probabilistic models of language. This paper contributes a sober view of the problem, a survey of techniques to address it, novel techniques, and extensions to the model. Our experiments on modelling written English text support a number of recommendations that should help researchers interested in this exciting field.
http://arxiv.org/abs/1904.08194
In object detection, keypoint-based approaches often suffer a large number of incorrect object bounding boxes, arguably due to the lack of an additional look into the cropped regions. This paper presents an efficient solution which explores the visual patterns within each cropped region with minimal costs. We build our framework upon a representative one-stage keypoint-based detector named CornerNet. Our approach, named CenterNet, detects each object as a triplet, rather than a pair, of keypoints, which improves both precision and recall. Accordingly, we design two customized modules named cascade corner pooling and center pooling, which play the roles of enriching information collected by both top-left and bottom-right corners and providing more recognizable information at the central regions, respectively. On the MS-COCO dataset, CenterNet achieves an AP of \textbf{47.0\%}, which outperforms all existing one-stage detectors by a large margin. Meanwhile, with a faster inference speed, CenterNet demonstrates quite comparable performance to the top-ranked two-stage detectors. Code is available at https://github.com/Duankaiwen/CenterNet.
http://arxiv.org/abs/1904.08189
Frequent mine disasters cause a large number of casualties and property losses. Autonomous driving is a fundamental measure for solving this problem, and track detection is one of the key technologies for computer vision to achieve downhole automatic driving. The track detection result based on the traditional convolutional neural network (CNN) algorithm lacks the detailed and unique description of the object and relies too much on visual postprocessing technology. Therefore, this paper proposes a track detection algorithm based on a multiscale conditional generative adversarial network (CGAN). The generator is decomposed into global and local parts using a multigranularity structure in the generator network. A multiscale shared convolution structure is adopted in the discriminator network to further supervise training the generator. Finally, the Monte Carlo search technique is introduced to search the intermediate state of the generator, and the result is sent to the discriminator for comparison. Compared with the existing work, our model achieved 82.43\% pixel accuracy and an average intersection-over-union (IOU) of 0.6218, and the detection of the track reached 95.01\% accuracy in the downhole roadway scene test set.
http://arxiv.org/abs/1904.08177
By their nature, the composition of black box models is opaque. This makes the ability to generate explanations for the response to stimuli challenging. The importance of explaining black box models has become increasingly important given the prevalence of AI and ML systems and the need to build legal and regulatory frameworks around them. Such explanations can also increase trust in these uncertain systems. In our paper we present RICE, a method for generating explanations of the behaviour of black box models by (1) probing a model to extract model output examples using sensitivity analysis; (2) applying CNPInduce, a method for inductive logic program synthesis, to generate logic programs based on critical input-output pairs; and (3) interpreting the target program as a human-readable explanation. We demonstrate the application of our method by generating explanations of an artificial neural network trained to follow simple traffic rules in a hypothetical self-driving car simulation. We conclude with a discussion on the scalability and usability of our approach and its potential applications to explanation-critical scenarios.
http://arxiv.org/abs/1904.09273
Objects in an image exhibit diverse scales. Adaptive receptive fields are expected to catch suitable range of context for accurate pixel level semantic prediction for handling objects of diverse sizes. Recently, atrous convolution with different dilation rates has been used to generate features of multi-scales through several branches and these features are fused for prediction. However, there is a lack of explicit interaction among the branches to adaptively make full use of the contexts. In this paper, we propose a Content-Adaptive Scale Interaction Network (CaseNet) to exploit the multi-scale features for scene parsing. We build the CaseNet based on the classic Atrous Spatial Pyramid Pooling (ASPP) module, followed by the proposed contextual scale interaction (CSI) module, and the scale adaptation (SA) module. Specifically, first, for each spatial position, we enable context interaction among different scales through scale-aware non-local operations across the scales, \ie, CSI module, which facilitates the generation of flexible mixed receptive fields, instead of a traditional flat one. Second, the scale adaptation module (SA) explicitly and softly selects the suitable scale for each spatial position and each channel. Ablation studies demonstrate the effectiveness of the proposed modules. We achieve state-of-the-art performance on three scene parsing benchmarks Cityscapes, ADE20K and LIP.
http://arxiv.org/abs/1904.08170
Artificial Neural Networks have shown impressive success in very different application cases. Choosing a proper network architecture is a critical decision for a network’s success, usually done in a manual manner. As a straightforward strategy, large, mostly fully connected architectures are selected, thereby relying on a good optimization strategy to find proper weights while at the same time avoiding overfitting. However, large parts of the final network are redundant. In the best case, large parts of the network become simply irrelevant for later inferencing. In the worst case, highly parameterized architectures hinder proper optimization and allow the easy creation of adverserial examples fooling the network. A first step in removing irrelevant architectural parts lies in identifying those parts, which requires measuring the contribution of individual components such as neurons. In previous work, heuristics based on using the weight distribution of a neuron as contribution measure have shown some success, but do not provide a proper theoretical understanding. Therefore, in our work we investigate game theoretic measures, namely the Shapley value (SV), in order to separate relevant from irrelevant parts of an artificial neural network. We begin by designing a coalitional game for an artificial neural network, where neurons form coalitions and the average contributions of neurons to coalitions yield to the Shapley value. In order to measure how well the Shapley value measures the contribution of individual neurons, we remove low-contributing neurons and measure its impact on the network performance. In our experiments we show that the Shapley value outperforms other heuristics for measuring the contribution of neurons.
http://arxiv.org/abs/1904.08166
In this study, we present an analysis of model-based ensemble learning for 3D point-cloud object classification and detection. An ensemble of multiple model instances is known to outperform a single model instance, but there is little study of the topic of ensemble learning for 3D point clouds. First, an ensemble of multiple model instances trained on the same part of the $\textit{ModelNet40}$ dataset was tested for seven deep learning, point cloud-based classification algorithms: $\textit{PointNet}$, $\textit{PointNet++}$, $\textit{SO-Net}$, $\textit{KCNet}$, $\textit{DeepSets}$, $\textit{DGCNN}$, and $\textit{PointCNN}$. Second, the ensemble of different architectures was tested. Results of our experiments show that the tested ensemble learning methods improve over state-of-the-art on the $\textit{ModelNet40}$ dataset, from $92.65\%$ to $93.64\%$ for the ensemble of single architecture instances, $94.03\%$ for two different architectures, and $94.15\%$ for five different architectures. We show that the ensemble of two models with different architectures can be as effective as the ensemble of 10 models with the same architecture. Third, a study on classic bagging i.e. with different subsets used for training multiple model instances) was tested and sources of ensemble accuracy growth were investigated for best-performing architecture, i.e. $\textit{SO-Net}$. We also investigate the ensemble learning of $\textit{Frustum PointNet}$ approach in the task of 3D object detection, increasing the average precision of 3D box detection on the $\textit{KITTI}$ dataset from $63.1\%$ to $66.5\%$ using only three model instances. We measure the inference time of all 3D classification architectures on a $\textit{Nvidia Jetson TX2}$, a common embedded computer for mobile robots, to allude to the use of these models in real-life applications.
http://arxiv.org/abs/1904.08159
In the first runtime analysis of an estimation-of-distribution algorithm (EDA) on the multi-modal jump function class, Hasen"ohrl and Sutton (GECCO 2018) proved that the runtime of the compact genetic algorithm with suitable parameter choice on jump functions with high probability is at most polynomial (in the dimension) if the jump size is at most logarithmic (in the dimension), and is at most exponential in the jump size if the jump size is super-logarithmic. The exponential runtime guarantee was achieved with a hypothetical population size that is also exponential in the jump size. Consequently, this setting cannot lead to a better runtime. In this work, we show that any choice of the hypothetical population size leads to a runtime that, with high probability, is at least exponential in the jump size. This result might be the first non-trivial exponential lower bound for EDAs that holds for arbitrary parameter settings.
http://arxiv.org/abs/1904.08415
This article reports on an investigation of the use of convolutional neural networks to predict the visual attention of chess players. The visual attention model described in this article has been created to generate saliency maps that capture hierarchical and spatial features of chessboard, in order to predict the probability fixation for individual pixels Using a skip-layer architecture of an autoencoder, with a unified decoder, we are able to use multiscale features to predict saliency of part of the board at different scales, showing multiple relations between pieces. We have used scan path and fixation data from players engaged in solving chess problems, to compute 6600 saliency maps associated to the corresponding chess piece configurations. This corpus is completed with synthetically generated data from actual games gathered from an online chess platform. Experiments realized using both scan-paths from chess players and the CAT2000 saliency dataset of natural images, highlights several results. Deep features, pretrained on natural images, were found to be helpful in training visual attention prediction for chess. The proposed neural network architecture is able to generate meaningful saliency maps on unseen chess configurations with good scores on standard metrics. This work provides a baseline for future work on visual attention prediction in similar contexts.
http://arxiv.org/abs/1904.08155
For many automated driving functions, a highly accurate perception of the vehicle environment is a crucial prerequisite. Modern high-resolution radar sensors generate multiple radar targets per object, which makes these sensors particularly suitable for the 2D object detection task. This work presents an approach to detect object hypotheses solely depending on sparse radar data using PointNets. In literature, only methods are presented so far which perform either object classification or bounding box estimation for objects. In contrast, this method facilitates a classification together with a bounding box estimation of objects using a single radar sensor. To this end, PointNets are adjusted for radar data performing 2D object classification with segmentation, and 2D bounding box regression in order to estimate an amodal bounding box. The algorithm is evaluated using an automatically created dataset which consist of various realistic driving maneuvers. The results show the great potential of object detection in high-resolution radar data using PointNets.
http://arxiv.org/abs/1904.08414
The recent advances in deep learning are mostly driven by availability of large amount of training data. However, availability of such data is not always possible for specific tasks such as speaker recognition where collection of large amount of data is not possible in practical scenarios. Therefore, in this paper, we propose to identify speakers by learning from only a few training examples. To achieve this, we use a deep neural network with prototypical loss where the input to the network is a spectrogram. For output, we project the class feature vectors into a common embedding space, followed by classification. Further, we show the effectiveness of capsule net in a few shot learning setting. To this end, we utilize an auto-encoder to learn generalized feature embeddings from class-specific embeddings obtained from capsule network. We provide exhaustive experiments on publicly available datasets and competitive baselines, demonstrating the superiority and generalization ability of the proposed few shot learning pipelines.
http://arxiv.org/abs/1904.08775
A major obstacle to developing artificial intelligence applications capable of true lifelong learning is that artificial neural networks quickly or catastrophically forget previously learned tasks when trained on a new one. Numerous methods for alleviating catastrophic forgetting are currently being proposed, but differences in evaluation protocols make it difficult to directly compare their performance. To enable more meaningful comparisons, here we identified three distinct scenarios for continual learning based on whether task identity is known and, if it is not, whether it needs to be inferred. Performing the split and permuted MNIST task protocols according to each of these scenarios, we found that regularization-based approaches (e.g., elastic weight consolidation) failed when task identity needed to be inferred. In contrast, generative replay combined with distillation (i.e., using class probabilities as “soft targets”) achieved superior performance in all three scenarios. Addressing the issue of efficiency, we reduced the computational cost of generative replay by integrating the generative model into the main model by equipping it with generative feedback or backward connections. This Replay-through-Feedback approach substantially shortened training time with no or negligible loss in performance. We believe this to be an important first step towards making the powerful technique of generative replay scalable to real-world continual learning applications.
http://arxiv.org/abs/1809.10635
Learning to take actions based on observations is a core requirement for artificial agents to be able to be successful and robust at their task. Reinforcement Learn-ing (RL) is a well-known technique for learning such policies. However, current RL algorithms often have to deal with reward shaping, have difficulties generalizing to other environments and are most often sample inefficient. In this paper, we explore active inference and the free energy principle, a normative theory from neuroscience that explains how self-organizing biological systems operate by maintaining a model of the world and casting action selection as an inference problem. We apply this concept to a typical problem known to the RL community, the mountain car problem, and show how active inference encompasses both RL and learning from demonstrations.
http://arxiv.org/abs/1904.08149
We address the problem of semi-supervised video object segmentation (VOS), where the masks of objects of interests are given in the first frame of an input video. To deal with challenging cases where objects are occluded or missing, previous work relies on greedy data association strategies that make decisions for each frame individually. In this paper, we propose a novel approach to defer the decision making for a target object in each frame, until a global view can be established with the entire video being taken into consideration. Our approach is in the same spirit as Multiple Hypotheses Tracking (MHT) methods, making several critical adaptations for the VOS problem. We employ the bounding box (bbox) hypothesis for tracking tree formation, and the multiple hypotheses are spawned by propagating the preceding bbox into the detected bbox proposals within a gated region starting from the initial object mask in the first frame. The gated region is determined by a gating scheme which takes into account a more comprehensive motion model rather than the simple Kalman filtering model in traditional MHT. To further design more customized algorithms tailored for VOS, we develop a novel mask propagation score instead of the appearance similarity score that could be brittle due to large deformations. The mask propagation score, together with the motion score, determines the affinity between the hypotheses during tree pruning. Finally, a novel mask merging strategy is employed to handle mask conflicts between objects. Extensive experiments on challenging datasets demonstrate the effectiveness of the proposed method, especially in the case of object missing.
http://arxiv.org/abs/1904.08141
Sentiment analysis research has been rapidly developing in the last decade and has attracted widespread attention from academia and industry, most of which is based on text. However, the information in the real world usually comes as different modalities. In this paper, we consider the task of Multimodal Sentiment Analysis, using Audio and Text Modalities, proposed a novel fusion strategy including Multi-Feature Fusion and Multi-Modality Fusion to improve the accuracy of Audio-Text Sentiment Analysis. We call this the Deep Feature Fusion-Audio and Text Modal Fusion (DFF-ATMF) model, and the features learned from it are complementary to each other and robust. Experiments with the CMU-MOSI corpus and the recently released CMU-MOSEI corpus for Youtube video sentiment analysis show the very competitive results of our proposed model. Surprisingly, our method also achieved the state-of-the-art results in the IEMOCAP dataset, indicating that our proposed fusion strategy is also extremely generalization ability to Multimodal Emotion Recognition.
http://arxiv.org/abs/1904.08138
Syntactic parsing is a key task in natural language processing. This task has been dominated by symbolic, grammar-based parsers. Neural networks, with their distributed representations, are challenging these methods. In this article we show that existing symbolic parsing algorithms can cross the border and be entirely formulated over distributed representations. To this end we introduce a version of the traditional Cocke-Younger-Kasami (CYK) algorithm, called D-CYK, which is entirely defined over distributed representations. Our D-CYK uses matrix multiplication on real number matrices of size independent of the length of the input string. These operations are compatible with traditional neural networks. Experiments show that our D-CYK approximates the original CYK algorithm. By showing that CYK can be entirely performed on distributed representations, we open the way to the definition of recurrent layers of CYK-informed neural networks.
http://arxiv.org/abs/1705.08843
Fueled by the diversity of datasets, semantic segmentation is a popular subfield in medical image analysis with a vast number of new methods being proposed each year. This ever-growing jungle of methodologies, however, becomes increasingly impenetrable. At the same time, many proposed methods fail to generalize beyond the experiments they were demonstrated on, thus hampering the process of developing a segmentation algorithm on a new dataset. Here we present nnU-Net (‘no-new-Net’), a framework that automatically adapts itself to any given new dataset. While this process was completely human-driven so far, we make a first attempt to automate necessary adaptations such as preprocessing, the exact patch size, batch size, and inference settings based on the properties of a given dataset. Remarkably, nnU-Net strips away the architectural bells and whistles that are typically proposed in the literature and relies on just a simple U-Net architecture embedded in a robust training scheme. Out of the box, nnU-Net achieves state of the art performance on six well-established segmentation challenges. Source code is available at https://github.com/MIC-DKFZ/nnunet.
http://arxiv.org/abs/1904.08128
This paper presents a taxonomy of explainability in Human-Agent Systems. We consider fundamental questions about the Why, Who, What, When and How of explainability. First, we define explainability, and its relationship to the related terms of interpretability, transparency, explicitness, and faithfulness. These definitions allow us to answer why explainability is needed in the system, whom it is geared to and what explanations can be generated to meet this need. We then consider when the user should be presented with this information. Last, we consider how objective and subjective measures can be used to evaluate the entire system. This last question is the most encompassing as it will need to evaluate all other issues regarding explainability.
http://arxiv.org/abs/1904.08123
In image restoration tasks, learning from discrete and fixed restoration levels, deep models cannot be easily generalized to data of continuous and unseen levels. We make a step forward by proposing a unified CNN framework that consists of few additional parameters than a single-level model yet could handle arbitrary restoration levels between a start and an end level. The additional module, namely AdaFM layer, performs channel-wise feature modification, and can adapt a model to another restoration level with high accuracy.
http://arxiv.org/abs/1904.08118
Traffic signal control is an important and challenging real-world problem, which aims to minimize the travel time of vehicles by coordinating their movements at the road intersections. Current traffic signal control systems in use still rely heavily on oversimplified information and rule-based methods, although we now have richer data, more computing power and advanced methods to drive the development of intelligent transportation. With the growing interest in intelligent transportation using machine learning methods like reinforcement learning, this survey covers the widely acknowledged transportation approaches and a comprehensive list of recent literature on reinforcement for traffic signal control. We hope this survey can foster interdisciplinary research on this important topic.
http://arxiv.org/abs/1904.08117
Dictionary learning methods can be split into two categories: i) class specific dictionary learning ii) class shared dictionary learning. The difference between the two categories is how to use the discriminative information. With the first category, samples of different classes are mapped to different subspaces which leads to some redundancy in the base vectors. For the second category, the samples in each specific class can not be described well. Moreover, most class shared dictionary learning methods use the L0-norm regularization term as the sparse constraint. In this paper, we first propose a novel class shared dictionary learning method named label embedded dictionary learning (LEDL) by introducing the L1-norm sparse constraint to replace the conventional L0-norm regularization term in LC-KSVD method. Then we propose a novel network named hybrid dictionary learning network (HDLN) to combine the class specific dictionary learning with class shared dictionary learning together to fully describe the feature to boost the performance of classification. Extensive experimental results on six benchmark datasets illustrate that our methods are capable of achieving superior performance compared to several conventional classification algorithms.
http://arxiv.org/abs/1904.08928