The image blurring process is generally modelled as the convolution of a blur kernel with a latent image. Therefore, the estimation of the blur kernel is essentially important for blind image deblurring. Unlike existing approaches which focus on approaching the problem by enforcing various priors on the blur kernel and the latent image, we are aiming at obtaining a high quality blur kernel directly by studying the problem in the frequency domain. We show that the auto-correlation of the absolute phase-only image can provide faithful information about the motion (e.g. the motion direction and magnitude, we call it the motion pattern in this paper.) that caused the blur, leading to a new and efficient blur kernel estimation approach. The blur kernel is then refined and the sharp image is estimated by solving an optimization problem by enforcing a regularization on the blur kernel and the latent image. We further extend our approach to handle non-uniform blur, which involves spatially varying blur kernels. Our approach is evaluated extensively on synthetic and real data and shows good results compared to the state-of-the-art deblurring approaches.
http://arxiv.org/abs/1811.10185
In this paper, we address the question answering challenge with the SQuAD 2.0 dataset. We design a model architecture which leverages BERT’s capability of context-aware word embeddings and BiDAF’s context interactive exploration mechanism. By integrating these two state-of-the-art architectures, our system tries to extract the contextual word representation at word and character levels, for better comprehension of both question and context and their correlations. We also propose our original joint posterior probability predictor module and its associated loss functions. Our best model so far obtains F1 score of 75.842% and EM score of 72.24% on the test PCE leaderboad.
http://arxiv.org/abs/1904.08109
Classical monocular vSLAM/VO methods suffer from the scale ambiguity problem. Hybrid approaches solve this problem by adding deep learning methods, for example by using depth maps which are predicted by a CNN. We suggest that it is better to base scale estimation on estimating the traveled distance for a set of subsequent images. In this paper, we propose a novel end-to-end many-to-one traveled distance estimator. By using a deep recurrent convolutional neural network (RCNN), the traveled distance between the first and last image of a set of consecutive frames is estimated by our DistanceNet. Geometric features are learned in the CNN part of our model, which are subsequently used by the RNN to learn dynamics and temporal information. Moreover, we exploit the natural order of distances by using ordinal regression to predict the distance. The evaluation on the KITTI dataset shows that our approach outperforms current state-of-the-art deep learning pose estimators and classical mono vSLAM/VO methods in terms of distance prediction. Thus, our DistanceNet can be used as a component to solve the scale problem and help improve current and future classical mono vSLAM/VO methods.
http://arxiv.org/abs/1904.08105
Recently, direct modeling of raw waveforms using deep neural networks has been widely studied for a number of tasks in audio domains. In speaker verification, however, utilization of raw waveforms is in its preliminary phase, requiring further investigation. In this study, we explore end-to-end deep neural networks that input raw waveforms to improve various aspects: front-end speaker embedding extraction including model architecture, pre-training scheme, additional objective functions, and back-end classification. Adjustment of model architecture using a pre-training scheme can extract speaker embeddings, giving a significant improvement in performance. Additional objective functions simplify the process of extracting speaker embeddings by merging conventional two-phase processes: extracting utterance-level features such as i-vectors or x-vectors and the feature enhancement phase, e.g., linear discriminant analysis. Effective back-end classification models that suit the proposed speaker embedding are also explored. We propose an end-to-end system that comprises two deep neural networks, one front-end for utterance-level speaker embedding extraction and the other for back-end classification. Experiments conducted on the VoxCeleb1 dataset demonstrate that the proposed model achieves state-of-the-art performance among systems without data augmentation. The proposed system is also comparable to the state-of-the-art x-vector system that adopts heavy data augmentation.
http://arxiv.org/abs/1904.08104
In this paper, we propose an efficient multi-scale geometric consistency guided multi-view stereo method for accurate and complete depth map estimation. We first present our basic multi-view stereo method with Adaptive Checkerboard sampling and Multi-Hypothesis joint view selection (ACMH). It leverages structured region information to sample better candidate hypotheses for propagation and infer the aggregation view subset at each pixel. For the depth estimation of low-textured areas, we further propose to combine ACMH with multi-scale geometric consistency guidance (ACMM) to obtain the reliable depth estimates for low-textured areas at coarser scales and guarantee that they can be propagated to finer scales. To correct the erroneous estimates propagated from the coarser scales, we present a novel detail restorer. Experiments on extensive datasets show our method achieves state-of-the-art performance, recovering the depth estimation not only in low-textured areas but also in details.
http://arxiv.org/abs/1904.08103
The number of approved patents worldwide increases rapidly each year, which requires new patent analytics to efficiently mine the valuable information attached to these patents. Vector space model (VSM) represents documents as high-dimensional vectors, where each dimension corresponds to a unique term. While originally proposed for information retrieval systems, VSM has also seen wide applications in patent analytics, and used as a fundamental tool to map patent documents to structured data. However, VSM method suffers from several limitations when applied to patent analysis tasks, such as loss of sentence-level semantics and curse-of-dimensionality problems. In order to address the above limitations, we propose a patent analytics based on feature vector space model (FVSM), where the FVSM is constructed by mapping patent documents to feature vectors extracted by convolutional neural networks (CNN). The applications of FVSM for three typical patent analysis tasks, i.e., patents similarity comparison, patent clustering, and patent map generation are discussed. A case study using patents related to Internet of Things (IoT) technology is illustrated to demonstrate the performance and effectiveness of FVSM. The proposed FVSM can be adopted by other patent analysis studies to replace VSM, based on which various big data learning tasks can be performed.
http://arxiv.org/abs/1904.08100
In this paper, we present correlated logistic (CorrLog) model for multilabel image classification. CorrLog extends conventional logistic regression model into multilabel cases, via explicitly modeling the pairwise correlation between labels. In addition, we propose to learn the model parameters of CorrLog with elastic net regularization, which helps exploit the sparsity in feature selection and label correlations and thus further boost the performance of multilabel classification. CorrLog can be efficiently learned, though approximately, by regularized maximum pseudo likelihood estimation, and it enjoys a satisfying generalization bound that is independent of the number of labels. CorrLog performs competitively for multilabel image classification on benchmark data sets MULAN scene, MIT outdoor scene, PASCAL VOC 2007, and PASCAL VOC 2012, compared with the state-of-the-art multilabel classification algorithms.
http://arxiv.org/abs/1904.08098
Many localized languages struggle to reap the benefits of recent advancements in character recognition systems due to the lack of substantial amount of labeled training data. This is due to the difficulty in generating large amounts of labeled data for such languages and inability of deep learning techniques to properly learn from small number of training samples. We solve this problem by introducing a technique of generating new training samples from the existing samples, with realistic augmentations which reflect actual variations that are present in human hand writing, by adding random controlled noise to their corresponding instantiation parameters. Our results with a mere 200 training samples per class surpass existing character recognition results in the EMNIST-letter dataset while achieving the existing results in the three datasets: EMNIST-balanced, EMNIST-digits, and MNIST. We also develop a strategy to effectively use a combination of loss functions to improve reconstructions. Our system is useful in character recognition for localized languages that lack much labeled training data and even in other related more general contexts such as object recognition.
http://arxiv.org/abs/1904.08095
Different layers of deep convolutional neural networks(CNNs) can encode different-level information. High-layer features always contain more semantic information, and low-layer features contain more detail information. However, low-layer features suffer from the background clutter and semantic ambiguity. During visual recognition, the feature combination of the low-layer and high-level features plays an important role in context modulation. If directly combining the high-layer and low-layer features, the background clutter and semantic ambiguity may be caused due to the introduction of detailed information. In this paper, we propose a general network architecture to concatenate CNN features of different layers in a simple and effective way, called Selective Feature Connection Mechanism (SFCM). Low-level features are selectively linked to high-level features with a feature selector which is generated by high-level features. The proposed connection mechanism can effectively overcome the above-mentioned drawbacks. We demonstrate the effectiveness, superiority, and universal applicability of this method on multiple challenging computer vision tasks, including image classification, scene text detection, and image-to-image translation.
http://arxiv.org/abs/1811.06295
Recently, researchers have started decomposing deep neural network models according to their semantics or functions. Recent work has shown the effectiveness of decomposed functional blocks for defending adversarial attacks, which add small input perturbation to the input image to fool the DNN models. This work proposes a profiling-based method to decompose the DNN models to different functional blocks, which lead to the effective path as a new approach to exploring DNNs’ internal organization. Specifically, the per-image effective path can be aggregated to the class-level effective path, through which we observe that adversarial images activate effective path different from normal images. We propose an effective path similarity-based method to detect adversarial images with an interpretable model, which achieve better accuracy and broader applicability than the state-of-the-art technique.
http://arxiv.org/abs/1904.08089
Bioimage classification plays a crucial role in many biological problems. Here we present a new General Purpose (GenP) ensemble that boosts performance by combining local features, dense sampling features, and deep learning approaches. We propose an ensemble of deep learning methods built using different criteria (different batch sizes, learning rates, topologies, and data augmentation methods). One of the contributions of this paper is the proposal of new methods of data augmentation based on feature transforms (principal component analysis/discrete cosine transform) that boost performance of Convolutional Neural Networks (CNNs). Each handcrafted descriptor is used to train a different Support Vector Machine (SVM), and the different SVMs are combined with the ensemble of CNNs. Our method is evaluated on a diverse set of bioimage classification problems. Results demonstrate that the proposed GenP bioimage ensemble obtains state-of-the-art performance without any ad-hoc dataset tuning of parameters (avoiding the risk of overfitting/overtraining).
http://arxiv.org/abs/1904.08084
We consider general discrete Markov Random Fields(MRFs) with additional bottleneck potentials which penalize the maximum (instead of the sum) over local potential value taken by the MRF-assignment. Bottleneck potentials or analogous constructions have been considered in (i) combinatorial optimization (e.g. bottleneck shortest path problem, the minimum bottleneck spanning tree problem, bottleneck function minimization in greedoids), (ii) inverse problems with $L_{\infty}$-norm regularization, and (iii) valued constraint satisfaction on the $(\min,\max)$-pre-semirings. Bottleneck potentials for general discrete MRFs are a natural generalization of the above direction of modeling work to Maximum-A-Posteriori (MAP) inference in MRFs. To this end, we propose MRFs whose objective consists of two parts: terms that factorize according to (i) $(\min,+)$, i.e. potentials as in plain MRFs, and (ii) $(\min,\max)$, i.e. bottleneck potentials. To solve the ensuing inference problem, we propose high-quality relaxations and efficient algorithms for solving them. We empirically show efficacy of our approach on large scale seismic horizon tracking problems.
http://arxiv.org/abs/1904.08080
Cluster analysis plays a very important role in data analysis. In these years, cluster ensemble, as a cluster analysis tool, has drawn much attention for its robustness, stability, and accuracy. Many efforts have been done to combine different initial clustering results into a single clustering solution with better performance. However, they neglect the structure information of the raw data in performing the cluster ensemble. In this paper, we propose a Structural Cluster Ensemble (SCE) algorithm for data partitioning formulated as a set-covering problem. In particular, we construct a Laplacian regularized objective function to capture the structure information among clusters. Moreover, considering the importance of the discriminative information underlying in the initial clustering results, we add a discriminative constraint into our proposed objective function. Finally, we verify the performance of the SCE algorithm on both synthetic and real data sets. The experimental results show the effectiveness of our proposed method SCE algorithm.
http://arxiv.org/abs/1904.08412
We propose the construction of a prototype scanner designed to capture multispectral images of documents. A standard sheet-feed scanner is modified by disconnecting its internal light source and connecting an external multispectral light source comprising of narrow band light emitting diodes (LED). A document is scanned by illuminating the scanner light guide successively with different LEDs and capturing a scan of the document. The system is portable and can be used for potential applications in verification of questioned documents, cheques, receipts and bank notes.
http://arxiv.org/abs/1904.12603
This paper studies panoptic segmentation, a recently proposed task which segments foreground (FG) objects at the instance level as well as background (BG) contents at the semantic level. Existing methods mostly dealt with these two problems separately, but in this paper, we reveal the underlying relationship between them, in particular, FG objects provide complementary cues to assist BG understanding. Our approach, named the Attention-guided Unified Network (AUNet), is a unified framework with two branches for FG and BG segmentation simultaneously. Two sources of attentions are added to the BG branch, namely, RPN and FG segmentation mask to provide object-level and pixel-level attentions, respectively. Our approach is generalized to different backbones with consistent accuracy gain in both FG and BG segmentation, and also sets new state-of-the-arts both in the MS-COCO (46.5% PQ) and Cityscapes (59.0% PQ) benchmarks.
http://arxiv.org/abs/1812.03904
End-to-end speech translation (ST), which directly translates from source language speech into target language text, has attracted intensive attentions in recent years. Compared to conventional pipeline systems, end-to-end ST models have advantages of lower latency, smaller model size and less error propagation. However, the combination of speech recognition and text translation in one model is more difficult than each of these two tasks. In this paper, we propose a knowledge distillation approach to improve ST model by transferring the knowledge from text translation model. Specifically, we first train a text translation model, regarded as a teacher model, and then ST model is trained to learn output probabilities from teacher model through knowledge distillation. Experiments on English- French Augmented LibriSpeech and English-Chinese TED corpus show that end-to-end ST is possible to implement on both similar and dissimilar language pairs. In addition, with the instruction of teacher model, end-to-end ST model can gain significant improvements by over 3.5 BLEU points.
http://arxiv.org/abs/1904.08075
Deep neural networks have been widely used in numerous computer vision applications, particularly in face recognition. However, deploying deep neural network face recognition on mobile devices has recently become a trend but still limited since most high-accuracy deep models are both time and GPU consumption in the inference stage. Therefore, developing a lightweight deep neural network is one of the most practical solutions to deploy face recognition on mobile devices. Such the lightweight deep neural network requires efficient memory with small number of weights representation and low cost operators. In this paper, a novel deep neural network named MobiFace, a simple but effective approach, is proposed for productively deploying face recognition on mobile devices. The experimental results have shown that our lightweight MobiFace is able to achieve high performance with 99.73% on LFW database and 91.3% on large-scale challenging Megaface database. It is also eventually competitive against large-scale deep-networks face recognition while significant reducing computational time and memory consumption.
http://arxiv.org/abs/1811.11080
In recent years, there has been an exponential growth in the number of complex documents and texts that require a deeper understanding of machine learning methods to be able to accurately classify texts in many applications. Many machine learning approaches have achieved surpassing results in natural language processing. The success of these learning algorithms relies on their capacity to understand complex models and non-linear relationships within data. However, finding suitable structures, architectures, and techniques for text classification is a challenge for researchers. In this paper, a brief overview of text classification algorithms is discussed. This overview covers different text feature extractions, dimensionality reduction methods, existing algorithms and techniques, and evaluations methods. Finally, the limitations of each technique and their application in the real-world problem are discussed.
http://arxiv.org/abs/1904.08067
The analysis of the collaborative learning process is one of the growing fields of education research, which has many different analytic solutions. In this paper, we provided a new solution to improve automated collaborative learning analyses using deep neural networks. Instead of using self-reported questionnaires, which are subject to bias and noise, we automatically extract group-working information by object recognition results using Mask R-CNN method. This process is based on detecting the people and other objects from pictures and video clips of the collaborative learning process, then evaluate the mobile learning performance using the collaborative indicators. We tested our approach to automatically evaluate the group-work collaboration in a controlled study of thirty-three dyads while performing an anatomy body painting intervention. The results indicate that our approach recognizes the differences of collaborations among teams of treatment and control groups in the case study. This work introduces new methods for automated quality prediction of collaborations among human-human interactions using computer vision techniques.
http://arxiv.org/abs/1904.08066
Feature-based time series representation has attracted substantial attention in a wide range of time series analysis methods. Recently, the use of time series features for forecast model selection and model averaging has been an emerging research focus in the forecasting community. Nonetheless, most of the existing approaches depend on the manual choice of an appropriate set of features. Exploiting machine learning methods to automatically extract features from time series becomes crucially important in the state-of-the-art time series analysis. In this paper, we introduce an automated approach to extract time series features based on images. Time series are first transformed into recurrence images, from which local features can be extracted using computer vision algorithms. The extracted features are used for forecast model selection and model averaging. Our experiments show that forecasting based on automatically extracted features, with less human intervention and a more comprehensive view of the raw time series data, yields comparable performances with the top best methods proposed in the largest forecasting competition M4.
http://arxiv.org/abs/1904.08064
We explore neural painters, a generative model for brushstrokes learned from a real non-differentiable and non-deterministic painting program. We show that when training an agent to “paint” images using brushstrokes, using a differentiable neural painter leads to much faster convergence. We propose a method for encouraging this agent to follow human-like strokes when reconstructing digits. We also explore the use of a neural painter as a differentiable image parameterization. By directly optimizing brushstrokes to activate neurons in a pre-trained convolutional network, we can directly visualize ImageNet categories and generate “ideal” paintings of each class. Finally, we present a new concept called intrinsic style transfer. By minimizing only the content loss from neural style transfer, we allow the artistic medium, in this case, brushstrokes, to naturally dictate the resulting style.
http://arxiv.org/abs/1904.08410
In recent years, the generation of conversation content based on deep neural networks has attracted many researchers. However, traditional neural language models tend to generate general replies, lacking logical and emotional factors. This paper proposes a conversation content generation model that combines reinforcement learning with emotional editing constraints to generate more meaningful and customizable emotional replies. The model divides the replies into three clauses based on pre-generated keywords and uses the emotional editor to further optimize the final reply. The model combines multi-task learning with multiple indicator rewards to comprehensively optimize the quality of replies. Experiments shows that our model can not only improve the fluency of the replies, but also significantly enhance the logical relevance and emotional relevance of the replies.
http://arxiv.org/abs/1904.08061
Deep image completion usually fails to harmonically blend the restored image into existing content, especially in the boundary area. This paper handles with this problem from a new perspective of creating a smooth transition and proposes a concise Deep Fusion Network (DFNet). Firstly, a fusion block is introduced to generate a flexible alpha composition map for combining known and unknown regions. The fusion block not only provides a smooth fusion between restored and existing content, but also provides an attention map to make network focus more on the unknown pixels. In this way, it builds a bridge for structural and texture information, so that information can be naturally propagated from known region into completion. Furthermore, fusion blocks are embedded into several decoder layers of the network. Accompanied by the adjustable loss constraints on each layer, more accurate structure information are achieved. We qualitatively and quantitatively compare our method with other state-of-the-art methods on Places2 and CelebA datasets. The results show the superior performance of DFNet, especially in the aspects of harmonious texture transition, texture detail and semantic structural consistency. Our source code will be avaiable at: \url{https://github.com/hughplay/DFNet}
http://arxiv.org/abs/1904.08060
Counting people or objects with significantly varying scales and densities has attracted much interest from the research community and yet it remains an open problem. In this paper, we propose a simple but an efficient and effective network, named DENet, which is composed of two components, i.e., a detection network (DNet) and an encoder-decoder estimation network (ENet). We first run DNet on an input image to detect and count individuals who can be segmented clearly. Then, ENet is utilized to estimate the density maps of the remaining areas, where the numbers of individuals cannot be detected. We propose a modified Xception as an encoder for feature extraction and a combination of dilated convolution and transposed convolution as a decoder. In the ShanghaiTech Part A, UCF and WorldExpo’10 datasets, our DENet achieves lower Mean Absolute Error (MAE) than those of the state-of-the-art methods.
http://arxiv.org/abs/1904.08056
Predicting the near-future from an input video is a useful task for applications such as autonomous driving and robotics. While most previous works predict a single future, multiple futures with different behaviors can possibly occur. Moreover, if the predicted future is too short, it may not be fully usable by a human or other system. In this paper, we propose a novel method for future video prediction capable of generating multiple long-term futures. This makes the predictions more suitable for real applications. First, from an input human video, we generate sequences of future human poses as the image coordinates of their body-joints by adversarial learning. We generate multiple futures by inputting to the generator combinations of a latent code (to reflect various behaviors) and an attraction point (to reflect various trajectories). In addition, we generate long-term future human poses using a novel approach based on unidimensional convolutional neural networks. Last, we generate an output video based on the generated poses for visualization. We evaluate the generated future poses and videos using three criteria (i.e., realism, diversity and accuracy), and show that our proposed method outperforms other state-of-the-art works.
https://arxiv.org/abs/1904.07538
This paper provides a new way to improve the efficiency of the REINFORCE training process. We apply it to the task of instance selection in distant supervision. Modeling the instance selection in one bag as a sequential decision process, a reinforcement learning agent is trained to determine whether an instance is valuable or not and construct a new bag with less noisy instances. However unbiased methods, such as REINFORCE, could usually take much time to train. This paper adopts posterior regularization (PR) to integrate some domain-specific rules in instance selection using REINFORCE. As the experiment results show, this method remarkably improves the performance of the relation classifier trained on cleaned distant supervision dataset as well as the efficiency of the REINFORCE training.
http://arxiv.org/abs/1904.08051
Recently, data-driven based Automatic Speech Recognition (ASR) systems have achieved state-of-the-art results. And transfer learning is often used when those existing systems are adapted to the target domain, e.g., fine-tuning, retraining. However, in the processes, the system parameters may well deviate too much from the previously learned parameters. Thus, it is difficult for the system training process to learn knowledge from target domains meanwhile not forgetting knowledge from the previous learning process, which is called as catastrophic forgetting (CF). In this paper, we attempt to solve the CF problem with the lifelong learning and propose a novel multi-task learning (MTL) training framework for ASR. It considers reserving original knowledge and learning new knowledge as two independent tasks, respectively. On the one hand, we constrain the new parameters not to deviate too far from the original parameters and punish the new system when forgetting original knowledge. On the other hand, we force the new system to solve new knowledge quickly. Then, a MTL mechanism is employed to get the balance between the two tasks. We applied our method to an End2End ASR task and obtained the best performance in both target and original datasets.
http://arxiv.org/abs/1904.08039
Machine learning has made major advances in categorizing objects in images, yet the best algorithms miss important aspects of how people learn and think about categories. People can learn richer concepts from fewer examples, including causal models that explain how members of a category are formed. Here, we explore the limits of this human ability to infer causal “programs” – latent generating processes with nontrivial algorithmic properties – from one, two, or three visual examples. People were asked to extrapolate the programs in several ways, for both classifying and generating new examples. As a theory of these inductive abilities, we present a Bayesian program learning model that searches the space of programs for the best explanation of the observations. Although variable, people’s judgments are broadly consistent with the model and inconsistent with several alternatives, including a pre-trained deep neural network for object recognition, indicating that people can learn and reason with rich algorithmic abstractions from sparse input data.
http://arxiv.org/abs/1904.08034
It is an effective way that improves the performance of the existing Automatic Speech Recognition (ASR) systems by retraining with more and more new training data in the target domain. Recently, Deep Neural Network (DNN) has become a successful model in the ASR field. In the training process of the DNN based methods, a back propagation of error between the transcription and the corresponding annotated text is used to update and optimize the parameters. Thus, the parameters are more influenced by the training samples with a big propagation error than the samples with a small one. In this paper, we define the samples with significant error as the hard samples and try to improve the performance of the ASR system by adding many of them. Unfortunately, the hard samples are sparse in the training data of the target domain, and manually label them is expensive. Therefore, we propose a hard samples mining method based on an enhanced deep multiple instance learning, which can find the hard samples from unlabeled training data by using a small subset of the dataset with manual labeling in the target domain. We applied our method to an End2End ASR task and obtained the best performance.
http://arxiv.org/abs/1904.08031
Describing what has changed in a scene can be useful to a user, but only if generated text focuses on what is semantically relevant. It is thus important to distinguish distractors (e.g. a viewpoint change) from relevant changes (e.g. an object has moved). We present a novel Dual Dynamic Attention Model (DUDA) to perform robust Change Captioning. Our model learns to distinguish distractors from semantic changes, localize the changes via Dual Attention over “before” and “after” images, and accurately describe them in natural language via Dynamic Speaker, by adaptively focusing on the necessary visual inputs (e.g. “before” or “after” image). To study the problem in depth, we collect a CLEVR-Change dataset, built off the CLEVR engine, with 5 types of scene changes. We benchmark a number of baselines on our dataset, and systematically study different change types and robustness to distractors. We show the superiority of our DUDA model in terms of both change captioning and localization. We also show that our approach is general, obtaining state-of-the-art results on the recent realistic Spot-the-Diff dataset which has no distractors.
https://arxiv.org/abs/1901.02527
Analyzing the geometric and semantic properties of 3D point clouds through the deep networks is still challenging due to the irregularity and sparsity of samplings of their geometric structures. This paper presents a new method to define and compute convolution directly on 3D point clouds by the proposed annular convolution. This new convolution operator can better capture the local neighborhood geometry of each point by specifying the (regular and dilated) ring-shaped structures and directions in the computation. It can adapt to the geometric variability and scalability at the signal processing level. We apply it to the developed hierarchical neural networks for object classification, part segmentation, and semantic segmentation in large-scale scenes. The extensive experiments and comparisons demonstrate that our approach outperforms the state-of-the-art methods on a variety of standard benchmark datasets (e.g., ModelNet10, ModelNet40, ShapeNet-part, S3DIS, and ScanNet).
http://arxiv.org/abs/1904.08017
This position paper presents a comparative study of co-occurrences. Some similarities and differences in the definition exist depending on the research domain (e.g. linguistics, NLP, computer science). This paper discusses these points, and deals with the methodological aspects in order to identify co-occurrences in a multidisciplinary paradigm.
http://arxiv.org/abs/1904.08010
Building a large image dataset with high-quality object masks for semantic segmentation is costly and time consuming. In this paper, we reduce the data preparation cost by leveraging weak supervision in the form of object bounding boxes. To accomplish this, we propose a principled framework that trains a deep convolutional segmentation model that combines a large set of weakly supervised images (having only object bounding box labels) with a small set of fully supervised images (having semantic segmentation labels and box labels). Our framework trains the primary segmentation model with the aid of an ancillary model that generates initial segmentation labels for the weakly supervised instances and a self-correction module that improves the generated labels during training using the increasingly accurate primary model. We introduce two variants of the self-correction module using either linear or convolutional functions. Experiments on the PASCAL VOC 2012 and Cityscape datasets show that our models trained with a small fully supervised set perform similar to, or better than, models trained with a large fully supervised set while requiring ~7x less annotation effort.
http://arxiv.org/abs/1811.07073
Detecting objects in aerial images is challenging for at least two reasons: (1) target objects like pedestrians are very small in terms of pixels, making them hard to be distinguished from surrounding background; and (2) targets are in general very sparsely and nonuniformly distributed, making the detection very inefficient. In this paper we address both issues inspired by the observation that these targets are often clustered. In particular, we propose a Clustered Detection (ClusDet) network that unifies object cluster and detection in an end-to-end framework. The key components in ClusDet include a cluster proposal sub-network (CPNet), a scale estimation sub-network (ScaleNet), and a dedicated detection network (DetecNet). Given an input image, CPNet produces (object) cluster regions and ScaleNet estimates object scales for these regions. Then, each scale-normalized cluster region and their features are fed into DetecNet for object detection. Compared with previous solutions, ClusDet has several advantages: (1) it greatly reduces the number of blocks for final object detection and hence achieves high running time efficiency, (2) the cluster-based scale estimation is more accurate than previously used single-object based ones, hence effectively improves the detection for small objects, and (3) the final DetecNet is dedicated for clustered regions and implicitly models the prior context information so as to boost detection accuracy. The proposed method is tested on three representative aerial image datasets including VisDrone, UAVDT and DOTA. In all the experiments, ClusDet achieves promising performance in both efficiency and accuracy, in comparison with state-of-the-art detectors.
http://arxiv.org/abs/1904.08008
This paper presents a formal definition and explicit representation of robot motion risk. Currently, robot motion risk has not been formally defined, but has already been used in motion and path planning. Risk is either implicitly represented as model uncertainty using probabilistic approaches, where the definition of risk is somewhat avoided, or explicitly modeled as a simple function of states, without a formal definition. In this work, we provide formal reasoning behind what risk is for robot motion and propose a formal definition of risk in terms of a sequence of motion, namely path. Mathematical approaches to represent motion risk are also presented, which is in accordance with our risk definition and properties. The definition and representation of risk provide a meaningful way to evaluate or construct robot motion or path plans. The understanding of risk is even of greater interest for the search and rescue community: the deconstructed environments cast extra risk onto the robot, since they are working under extreme conditions. A proper risk representation has the potential to reduce robot failure in the field.
http://arxiv.org/abs/1904.08003
Object detection in still images has drawn a lot of attention over past few years, and with the advent of Deep Learning impressive performances have been achieved with numerous industrial applications. Most of these deep learning models rely on RGB images to localize and identify objects in the image. However in some application scenarii, images are compressed either for storage savings or fast transmission. Therefore a time consuming image decompression step is compulsory in order to apply the aforementioned deep models. To alleviate this drawback, we propose a fast deep architecture for object detection in JPEG images, one of the most widespread compression format. We train a neural network to detect objects based on the blockwise DCT (discrete cosine transform) coefficients {issued from} the JPEG compression algorithm. We modify the well-known Single Shot multibox Detector (SSD) by replacing its first layers with one convolutional layer dedicated to process the DCT inputs. Experimental evaluations on PASCAL VOC and industrial dataset comprising images of road traffic surveillance show that the model is about $2\times$ faster than regular SSD with promising detection performances. To the best of our knowledge, this paper is the first to address detection in compressed JPEG images.
http://arxiv.org/abs/1904.08408
This paper proposes and benchmarks two tether-based motion primitives for tethered UAVs to execute autonomous flight with proprioception only. Tethered UAVs have been studied mainly due to power and safety considerations. Tether is either not included in the UAV motion (treated same as free-flying UAV) or only in terms of station-keeping and high-speed steady flight. However, feedback from and control over the tether configuration could be utilized as a set of navigational tools for autonomous flight, especially in GPS-denied environments and without vision-based exteroception. In this work, two tether-based motion primitives are proposed, which can enable autonomous flight of a tethered UAV. The proposed motion primitives are implemented on a physical tethered UAV for autonomous path execution with motion capture ground truth. The navigational performance is quantified and compared. The proposed motion primitives make tethered UAV a mobile and safe autonomous robot platform. The benchmarking results suggest appropriate usage of the two motion primitives for tethered UAVs with different path plans.
http://arxiv.org/abs/1904.07996
The use of subword-level information (e.g., characters, character n-grams, morphemes) has become ubiquitous in modern word representation learning. Its importance is attested especially for morphologically rich languages which generate a large number of rare words. Despite a steadily increasing interest in such subword-informed word representations, their systematic comparative analysis across typologically diverse languages and different tasks is still missing. In this work, we deliver such a study focusing on the variation of two crucial components required for subword-level integration into word representation models: 1) segmentation of words into subword units, and 2) subword composition functions to obtain final word representations. We propose a general framework for learning subword-informed word representations that allows for easy experimentation with different segmentation and composition components, also including more advanced techniques based on position embeddings and self-attention. Using the unified framework, we run experiments over a large number of subword-informed word representation configurations (60 in total) on 3 tasks (general and rare word similarity, dependency parsing, fine-grained entity typing) for 5 languages representing 3 language types. Our main results clearly indicate that there is no “one-sizefits-all” configuration, as performance is both language- and task-dependent. We also show that configurations based on unsupervised segmentation (e.g., BPE, Morfessor) are sometimes comparable to or even outperform the ones based on supervised word segmentation.
http://arxiv.org/abs/1904.07994
Community question-answering (CQA) platforms have become very popular forums for asking and answering questions daily. While these forums are rich repositories of community knowledge, they present challenges for finding relevant answers and similar questions, due to the open-ended nature of informal discussions. Further, if the platform allows questions and answers in multiple languages, we are faced with the additional challenge of matching cross-lingual information. In this work, we focus on the cross-language question re-ranking shared task, which aims to find existing questions that may be written in different languages. Our contribution is an exploration of query expansion techniques for this problem. We investigate expansions based on Word Embeddings, DBpedia concepts linking, and Hypernym, and show that they outperform existing state-of-the-art methods.
http://arxiv.org/abs/1904.07982
Reinforcement Learning (RL) algorithms allow artificial agents to improve their action selections so as to increase rewarding experiences in their environments. Deep Reinforcement Learning algorithms require solving a nonconvex and nonlinear unconstrained optimization problem. Methods for solving the optimization problems in deep RL are restricted to the class of first-order algorithms, such as stochastic gradient descent (SGD). The major drawback of the SGD methods is that they have the undesirable effect of not escaping saddle points and their performance can be seriously obstructed by ill-conditioning. Furthermore, SGD methods require exhaustive trial and error to fine-tune many learning parameters. Using second derivative information can result in improved convergence properties, but computing the Hessian matrix for large-scale problems is not practical. Quasi-Newton methods require only first-order gradient information, like SGD, but they can construct a low rank approximation of the Hessian matrix and result in superlinear convergence. The limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) approach is one of the most popular quasi-Newton methods that construct positive definite Hessian approximations. In this paper, we introduce an efficient optimization method, based on the limited memory BFGS quasi-Newton method using line search strategy – as an alternative to SGD methods. Our method bridges the disparity between first order methods and second order methods by continuing to use gradient information to calculate a low-rank Hessian approximations. We provide formal convergence analysis as well as empirical results on a subset of the classic ATARI 2600 games. Our results show a robust convergence with preferred generalization characteristics, as well as fast training time and no need for the experience replaying mechanism.
http://arxiv.org/abs/1811.02693
We analyze clouds in the earth’s atmosphere using ground-based sky cameras. An accurate segmentation of clouds in the captured sky/cloud image is difficult, owing to the fuzzy boundaries of clouds. Several techniques have been proposed that use color as the discriminatory feature for cloud detection. In the existing literature, however, analysis of daytime and nighttime images is considered separately, mainly because of differences in image characteristics and applications. In this paper, we propose a light-weight deep-learning architecture called CloudSegNet. It is the first that integrates daytime and nighttime (also known as nychthemeron) image segmentation in a single framework, and achieves state-of-the-art results on public databases.
http://arxiv.org/abs/1904.07979
Is cognition a collection of loosely connected functions tuned to different tasks, or can there be a general learning algorithm? If such an hypothetical general algorithm did exist, tuned to our world, could it adapt seamlessly to a world with different laws of nature? We consider the theory that predictive coding is such a general rule, and falsify it for one specific neural architecture known for high-performance predictions on natural videos and replication of human visual illusions: PredNet. Our results show that PredNet’s high performance generalizes without retraining on a completely different natural video dataset. Yet PredNet cannot be trained to reach even mediocre accuracy on an artificial video dataset created with the rules of the Game of Life (GoL). We also find that a submodule of PredNet, a Convolutional Neural Network trained alone, reaches perfect accuracy on the GoL while being mediocre for natural videos, showing that PredNet’s architecture itself is responsible for both the high performance on natural videos and the loss of performance on the GoL. Just as humans cannot predict the dynamics of the GoL, our results suggest that there might be a trade-off between high performance on sensory inputs with different sets of rules.
http://arxiv.org/abs/1904.07969
Visual Place Recognition (VPR) has seen significant advances at the frontiers of matching performance and computational superiority over the past few years. However, these evaluations are performed for ground-based mobile platforms and cannot be generalized to aerial platforms. The degree of viewpoint variation experienced by aerial robots is complex, with their processing power and on-board memory limited by payload size and battery ratings. Therefore, in this paper, we collect $8$ state-of-the-art VPR techniques that have been previously evaluated for ground-based platforms and compare them on $2$ recently proposed aerial place recognition datasets with three prime focuses: a) Matching performance b) Processing power consumption c) Projected memory requirements. This gives a birds-eye view of the applicability of contemporary VPR research to aerial robotics and lays down the the nature of challenges for aerial-VPR.
http://arxiv.org/abs/1904.07967
Corporate mobility is often based on a fixed assignment of vehicles to employees. Relaxing this fixation while including alternatives such as public transportation or taxis for business and private trips could increase fleet utilization and foster the use of battery electric vehicles. Along this idea we propose a flexible booking system, leading to the introduction of the NP-hard mobility offer allocation problem which is closely related to multi-interval scheduling problems. We describe problem specific conflict graphs for representing and exploring the structure of feasible solutions. A characterization of all maximum cliques in these conflict graphs reveals symmetries which allow to formulate stronger integer linear programming models. We also present an adaptive large neighborhood search based approach which makes use of conflict graphs as well. In a computational study, the approaches are evaluated and it is demonstrated that, depending on instances and run-time requirements, either a solver for the integer linear programming model, fast greedy heuristics, or the adaptive large neighborhood search outperforms the others.
http://arxiv.org/abs/1810.05659
Natural language processing tools are used to automatically detect disturbances in transcribed speech of schizophrenia inpatients who speak Hebrew. We measure topic mutation over time and show that controls maintain more cohesive speech than inpatients. We also examine differences in how inpatients and controls use adjectives and adverbs to describe content words and show that the ones used by controls are more common than the those of inpatients. We provide experimental results and show their potential for automatically detecting schizophrenia in patients by means only of their speech patterns.
http://arxiv.org/abs/1904.07953
A plethora of deep learning models have been developed for the task of Alzheimer’s disease classification from brain MRI scans. Many of these models report high performance, achieving three-class classification accuracy of up to 95%. However, it is common for these studies to draw performance comparisons between models that are trained on different subsets of a dataset or use varying imaging preprocessing techniques, making it difficult to objectively assess model performance. Furthermore, many of these works do not provide details such as hyperparameters, the specific MRI scans used, or their source code, making it difficult to replicate their experiments. To address these concerns, we present a comprehensive study of some of the deep learning methods and architectures on the full set of images available from ADNI. We find that, (1) classification using 3D models gives an improvement of 1% in our setup, at the cost of significantly longer training time and more computation power, (2) with our dataset, pre-training yields minimal ($<0.5\%$) improvement in model performance, (3) most popular convolutional neural network models yield similar performance when compared to each other. Lastly, we briefly compare the effects of two image preprocessing programs: FreeSurfer and Clinica, and find that the spatially normalized and segmented outputs from Clinica increased the accuracy of model prediction from 63% to 89% when compared to FreeSurfer images.
http://arxiv.org/abs/1904.07950
Recent approaches in text-to-speech (TTS) synthesis employ neural network strategies to vocode perceptually-informed spectrogram representations directly into listenable waveforms. Such vocoding procedures create a computational bottleneck in modern TTS pipelines. We propose an alternative approach which utilizes generative adversarial networks (GANs) to learn mappings from perceptually-informed spectrograms to simple magnitude spectrograms which can be heuristically vocoded. Through a user study, we show that our approach significantly outperforms na"ive vocoding strategies while being hundreds of times faster than neural network vocoders used in state-of-the-art TTS systems. We also show that our method can be used to achieve state-of-the-art results in unsupervised synthesis of individual words of speech.
http://arxiv.org/abs/1904.07944
Similarity search is a popular technique for seismic signal processing, with template matching, matched filters and subspace detectors being utilized for a wide variety of tasks, including both signal detection and source discrimination. Traditionally, these techniques rely on the cross-correlation function as the basis for measuring similarity. Unfortunately, seismogram correlation is dominated by path effects, essentially requiring a distinct waveform template along each path to be detected. To address this limitation, we define a path-invariant measure for seismogram similarity. A deep convolutional neural network with a triplet loss function maps raw seismograms to a low dimensional embedding space, where nearness on the space corresponds to nearness of source function, regardless of path or recording instrumentation. This path-agnostic embedding space represents a new representation for seismograms, characterized by robust, source-specific features. The dataset used to train and test the algorithm comes primarily from the USArray experiment, a temporary network of 400 seismometers that was deployed at more than 2000 locations across the US from 2007 to 2012. The first four years (2006, 2007, 2008, 2009) were selected as the training set, and the following two years (2010, 2011) were selected for validation and testing. The training, validation and test sets contained 24,811, 5,711 and 4,214 seismograms, respectively. The utility of our novel embedding space representation is evaluated across three common seismic tasks: event association, signal detection, and source discrimination, achieving an accuracy of 80%, 92% and 90%, respectively, all while minimizing the number of template waveforms required.
http://arxiv.org/abs/1904.07936
We tackle the problem of semantic boundary prediction, which aims to identify pixels that belong to object(class) boundaries. We notice that relevant datasets consist of a significant level of label noise, reflecting the fact that precise annotations are laborious to get and thus annotators trade-off quality with efficiency. We aim to learn sharp and precise semantic boundaries by explicitly reasoning about annotation noise during training. We propose a simple new layer and loss that can be used with existing learning-based boundary detectors. Our layer/loss enforces the detector to predict a maximum response along the normal direction at an edge, while also regularizing its direction. We further reason about true object boundaries during training using a level set formulation, which allows the network to learn from misaligned labels in an end-to-end fashion. Experiments show that we improve over the CASENet backbone network by more than 4% in terms of MF(ODS) and 18.61% in terms of AP, outperforming all current state-of-the-art methods including those that deal with alignment. Furthermore, we show that our learned network can be used to significantly improve coarse segmentation labels, lending itself as an efficient way to label new data.
http://arxiv.org/abs/1904.07934
In this paper, we investigate how to learn rich and robust feature representations for audio classification from visual data and a novel audio data modality, namely acoustic images. Former models learn audio representations from raw signals or spectral data acquired by a single microphone, with remarkable results in classification and retrieval. However, such representations are not so robust towards variable environmental sound conditions. We tackle this drawback by exploiting a new multimodal labeled action recognition dataset acquired by a hybrid audio-visual sensor that provides RGB video, raw audio signals, and spatialized acoustic data, also known as acoustic images, where the visual and acoustic images are aligned in space and synchronized in time. Using this richer information, we train audio deep learning models in a teacher-student fashion. In particular, we distill knowledge into audio networks from both visual and acoustic image teachers. Our experiments suggest that the learned representations are more powerful and have better generalization capabilities than the features learned from models trained using just visual or single-microphone audio data.
http://arxiv.org/abs/1904.07933