PyTorchKR Papers

최신 논문

총 1,314편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1703.09167

A Study on the Extraction and Analysis of a Large Set of Eye Movement Features during Reading

This work presents a study on the extraction and analysis of a set of 101 categories of eye movement features from three types of eye movement events: fixations, saccades, and post-saccadic oscillations. The eye movements were recorded during a reading task. For the categories of features with multiple instances in a recording we extract corresponding feature subtypes by calculating descriptive statistics on the distributions of these instances. A unified framework of detailed descriptions and mathematical formulas are provided for the extraction of the feature set. The analysis of feature values is performed using a large database of eye movement recordings from a normative population of 298 subjects. We demonstrate the central tendency and overall variability of feature values over the experimental population, and more importantly, we quantify the test-retest reliability (repeatability) of each separate feature. The described methods and analysis can provide valuable tools in fields exploring the eye movements, such as in behavioral studies, attention and cognition research, medical research, biometric recognition, and human-computer interaction.

arXiv:1702.07191

ViP-CNN: Visual Phrase Guided Convolutional Neural Network

As the intermediate level task connecting image captioning and object detection, visual relationship detection started to catch researchers' attention because of its descriptive power and clear structure. It detects the objects and captures their pair-wise interactions with a subject-predicate-object triplet, e.g. person-ride-horse. In this paper, each visual relationship is considered as a phrase with three components. We formulate the visual relationship detection as three inter-connected recognition problems and propose a Visual Phrase guided Convolutional Neural Network (ViP-CNN) to address them simultaneously. In ViP-CNN, we present a Phrase-guided Message Passing Structure (PMPS) to establish the connection among relationship components and help the model consider the three problems jointly. Corresponding non-maximum suppression method and model training strategy are also proposed. Experimental results show that our ViP-CNN outperforms the state-of-art method both in speed and accuracy. We further pretrain ViP-CNN on our cleansed Visual Genome Relationship dataset, which is found to perform better than the pretraining on the ImageNet for this task.

arXiv:1611.06607

A Hierarchical Approach for Generating Descriptive Image Paragraphs

Recent progress on image captioning has made it possible to generate novel sentences describing images in natural language, but compressing an image into a single sentence can describe visual content in only coarse detail. While one new captioning approach, dense captioning, can potentially describe images in finer levels of detail by captioning many regions within an image, it in turn is unable to produce a coherent story for an image. In this paper we overcome these limitations by generating entire paragraphs for describing images, which can tell detailed, unified stories. We develop a model that decomposes both images and paragraphs into their constituent parts, detecting semantic regions in images and using a hierarchical recurrent neural network to reason about language. Linguistic analysis confirms the complexity of the paragraph generation task, and thorough experiments on a new dataset of image and paragraph pairs demonstrate the effectiveness of our approach.

arXiv:1704.07427

Recognizing Descriptive Wikipedia Categories for Historical Figures

Wikipedia is a useful knowledge source that benefits many applications in language processing and knowledge representation. An important feature of Wikipedia is that of categories. Wikipedia pages are assigned different categories according to their contents as human-annotated labels which can be used in information retrieval, ad hoc search improvements, entity ranking and tag recommendations. However, important pages are usually assigned too many categories, which makes it difficult to recognize the most important ones that give the best descriptions. In this paper, we propose an approach to recognize the most descriptive Wikipedia categories. We observe that historical figures in a precise category presumably are mutually similar and such categorical coherence could be evaluated via texts or Wikipedia links of corresponding members in the category. We rank descriptive level of Wikipedia categories according to their coherence and our ranking yield an overall agreement of 88.27% compared with human wisdom.

arXiv:1704.06485공식 코드 · Papers with Code

Attend to You: Personalized Image Captioning with Context Sequence Memory Networks

We address personalization issues of image captioning, which have not been discussed yet in previous research. For a query image, we aim to generate a descriptive sentence, accounting for prior knowledge such as the user's active vocabularies in previous documents. As applications of personalized image captioning, we tackle two post automation tasks: hashtag prediction and post generation, on our newly collected Instagram dataset, consisting of 1.1M posts from 6.3K users. We propose a novel captioning model named Context Sequence Memory Network (CSMN). Its unique updates over previous memory network models include (i) exploiting memory as a repository for multiple types of context information, (ii) appending previously generated words into memory to capture long-term information without suffering from the vanishing gradient problem, and (iii) adopting CNN memory structure to jointly represent nearby ordered memory slots for better context understanding. With quantitative evaluation and user studies via Amazon Mechanical Turk, we show the effectiveness of the three novel features of CSMN and its performance enhancement for personalized image captioning over state-of-the-art captioning models.

arXiv:1705.02460

Image Annotation using Multi-Layer Sparse Coding

Automatic annotation of images with descriptive words is a challenging problem with vast applications in the areas of image search and retrieval. This problem can be viewed as a label-assignment problem by a classifier dealing with a very large set of labels, i.e., the vocabulary set. We propose a novel annotation method that employs two layers of sparse coding and performs coarse-to-fine labeling. Themes extracted from the training data are treated as coarse labels. Each theme is a set of training images that share a common subject in their visual and textual contents. Our system extracts coarse labels for training and test images without requiring any prior knowledge. Vocabulary words are the fine labels to be associated with images. Most of the annotation methods achieve low recall due to the large number of available fine labels, i.e., vocabulary words. These systems also tend to achieve high precision for highly frequent words only while relatively rare words are more important for search and retrieval purposes. Our system not only outperforms various previously proposed annotation systems, but also achieves symmetric response in terms of precision and recall. Our system scores and maintains high precision for words with a wide range of frequencies. Such behavior is achieved by intelligently reducing the number of available fine labels or words for each image based on coarse labels assigned to it.

arXiv:1705.03669

Mind the Gap: A Well Log Data Analysis

The main task in oil and gas exploration is to gain an understanding of the distribution and nature of rocks and fluids in the subsurface. Well logs are records of petro-physical data acquired along a borehole, providing direct information about what is in the subsurface. The data collected by logging wells can have significant economic consequences, due to the costs inherent to drilling wells, and the potential return of oil deposits. In this paper, we describe preliminary work aimed at building a general framework for well log prediction. First, we perform a descriptive and exploratory analysis of the gaps in the neutron porosity logs of more than a thousand wells in the North Sea. Then, we generate artificial gaps in the neutron logs that reflect the statistics collected before. Finally, we compare Artificial Neural Networks, Random Forests, and three algorithms of Linear Regression in the prediction of missing gaps on a well-by-well basis.

arXiv:1705.04662공식 코드 · Papers with Code

Monaural Audio Speaker Separation with Source Contrastive Estimation

We propose an algorithm to separate simultaneously speaking persons from each other, the "cocktail party problem", using a single microphone. Our approach involves a deep recurrent neural networks regression to a vector space that is descriptive of independent speakers. Such a vector space can embed empirically determined speaker characteristics and is optimized by distinguishing between speaker masks. We call this technique source-contrastive estimation. The methodology is inspired by negative sampling, which has seen success in natural language processing, where an embedding is learned by correlating and de-correlating a given input vector with output weights. Although the matrix determined by the output weights is dependent on a set of known speakers, we only use the input vectors during inference. Doing so will ensure that source separation is explicitly speaker-independent. Our approach is similar to recent deep neural network clustering and permutation-invariant training research; we use weighted spectral features and masks to augment individual speaker frequencies while filtering out other speakers. We avoid, however, the severe computational burden of other approaches with our technique. Furthermore, by training a vector space rather than combinations of different speakers or differences thereof, we avoid the so-called permutation problem during training. Our algorithm offers an intuitive, computationally efficient response to the cocktail party problem, and most importantly boasts better empirical performance than other current techniques.

arXiv:1706.00130

Teaching Machines to Describe Images via Natural Language Feedback

Robots will eventually be part of every household. It is thus critical to enable algorithms to learn from and be guided by non-expert users. In this paper, we bring a human in the loop, and enable a human teacher to give feedback to a learning agent in the form of natural language. We argue that a descriptive sentence can provide a much stronger learning signal than a numeric reward in that it can easily point to where the mistakes are and how to correct them. We focus on the problem of image captioning in which the quality of the output can easily be judged by non-experts. We propose a hierarchical phrase-based captioning model trained with policy gradients, and design a feedback network that provides reward to the learner by conditioning on the human-provided feedback. We show that by exploiting descriptive feedback our model learns to perform better than when given independently written human captions.

arXiv:1606.05729

RRV: A Spatiotemporal Descriptor for Rigid Body Motion Recognition

Motion behaviors of a rigid body can be characterized by a 6-dimensional motion trajectory, which contains position vectors of a reference point on the rigid body and rotations of this rigid body over time. This paper devises a Rotation and Relative Velocity (RRV) descriptor by exploring the local translational and rotational invariants of motion trajectories of rigid bodies, which is insensitive to noise, invariant to rigid transformation and scaling. A flexible metric is also introduced to measure the distance between two RRV descriptors. The RRV descriptor is then applied to characterize motions of a human body skeleton modeled as articulated interconnections of multiple rigid bodies. To illustrate the descriptive ability of the RRV descriptor, we explore it for different rigid body motion recognition tasks. The experimental results on benchmark datasets demonstrate that this simple RRV descriptor outperforms the previous ones regarding recognition accuracy without increasing computational cost.

arXiv:1706.04388

Alignment Distances on Systems of Bags

Recent research in image and video recognition indicates that many visual processes can be thought of as being generated by a time-varying generative model. A nearby descriptive model for visual processes is thus a statistical distribution that varies over time. Specifically, modeling visual processes as streams of histograms generated by a kernelized linear dynamic system turns out to be efficient. We refer to such a model as a System of Bags. In this work, we investigate Systems of Bags with special emphasis on dynamic scenes and dynamic textures. Parameters of linear dynamic systems suffer from ambiguities. In order to cope with these ambiguities in the kernelized setting, we develop a kernelized version of the alignment distance. For its computation, we use a Jacobi-type method and prove its convergence to a set of critical points. We employ it as a dissimilarity measure on Systems of Bags. As such, it outperforms other known dissimilarity measures for kernelized linear dynamic systems, in particular the Martin Distance and the Maximum Singular Value Distance, in every tested classification setting. A considerable margin can be observed in settings, where classification is performed with respect to an abstract mean of video sets. For this scenario, the presented approach can outperform state-of-the-art techniques, such as Dynamic Fractal Spectrum or Orthogonal Tensor Dictionary Learning.

arXiv:1707.00297

Classification non supervisée des données hétérogènes à large échelle

When it comes to cluster massive data, response time, disk access and quality of formed classes becoming major issues for companies. It is in this context that we have come to define a clustering framework for large scale heterogeneous data that contributes to the resolution of these issues. The proposed framework is based on, firstly, the descriptive analysis based on MCA, and secondly, the MapReduce paradigm in a large scale environment. The results are encouraging and prove the efficiency of the hybrid deployment on response quality and time component as on qualitative and quantitative data.

arXiv:1701.01879공식 코드 · Papers with Code

Greedy Search for Descriptive Spatial Face Features

Facial expression recognition methods use a combination of geometric and appearance-based features. Spatial features are derived from displacements of facial landmarks, and carry geometric information. These features are either selected based on prior knowledge, or dimension-reduced from a large pool. In this study, we produce a large number of potential spatial features using two combinations of facial landmarks. Among these, we search for a descriptive subset of features using sequential forward selection. The chosen feature subset is used to classify facial expressions in the extended Cohn-Kanade dataset (CK+), and delivered 88.7% recognition accuracy without using any appearance-based features.

arXiv:1705.03566

Spatial Random Sampling: A Structure-Preserving Data Sketching Tool

Random column sampling is not guaranteed to yield data sketches that preserve the underlying structures of the data and may not sample sufficiently from less-populated data clusters. Also, adaptive sampling can often provide accurate low rank approximations, yet may fall short of producing descriptive data sketches, especially when the cluster centers are linearly dependent. Motivated by that, this paper introduces a novel randomized column sampling tool dubbed Spatial Random Sampling (SRS), in which data points are sampled based on their proximity to randomly sampled points on the unit sphere. The most compelling feature of SRS is that the corresponding probability of sampling from a given data cluster is proportional to the surface area the cluster occupies on the unit sphere, independently from the size of the cluster population. Although it is fully randomized, SRS is shown to provide descriptive and balanced data representations. The proposed idea addresses a pressing need in data science and holds potential to inspire many novel approaches for analysis of big data.

arXiv:1707.06029

Supervising Neural Attention Models for Video Captioning by Human Gaze Data

The attention mechanisms in deep neural networks are inspired by human's attention that sequentially focuses on the most relevant parts of the information over time to generate prediction output. The attention parameters in those models are implicitly trained in an end-to-end manner, yet there have been few trials to explicitly incorporate human gaze tracking to supervise the attention models. In this paper, we investigate whether attention models can benefit from explicit human gaze labels, especially for the task of video captioning. We collect a new dataset called VAS, consisting of movie clips, and corresponding multiple descriptive sentences along with human gaze tracking data. We propose a video captioning model named Gaze Encoding Attention Network (GEAN) that can leverage gaze tracking information to provide the spatial and temporal attention for sentence generation. Through evaluation of language similarity metrics and human assessment via Amazon mechanical Turk, we demonstrate that spatial attentions guided by human gaze data indeed improve the performance of multiple captioning methods. Moreover, we show that the proposed approach achieves the state-of-the-art performance for both gaze prediction and video captioning not only in our VAS dataset but also in standard datasets (e.g. LSMDC and Hollywood2).

arXiv:1707.07673

Evaluation of Semantic Web Technologies for Storing Computable Definitions of Electronic Health Records Phenotyping Algorithms

Electronic Health Records are electronic data generated during or as a byproduct of routine patient care. Structured, semi-structured and unstructured EHR offer researchers unprecedented phenotypic breadth and depth and have the potential to accelerate the development of precision medicine approaches at scale. A main EHR use-case is defining phenotyping algorithms that identify disease status, onset and severity. Phenotyping algorithms utilize diagnoses, prescriptions, laboratory tests, symptoms and other elements in order to identify patients with or without a specific trait. No common standardized, structured, computable format exists for storing phenotyping algorithms. The majority of algorithms are stored as human-readable descriptive text documents making their translation to code challenging due to their inherent complexity and hinders their sharing and re-use across the community. In this paper, we evaluate the two key Semantic Web Technologies, the Web Ontology Language and the Resource Description Framework, for enabling computable representations of EHR-driven phenotyping algorithms.

arXiv:1707.07102

OBJ2TEXT: Generating Visually Descriptive Language from Object Layouts

Generating captions for images is a task that has recently received considerable attention. In this work we focus on caption generation for abstract scenes, or object layouts where the only information provided is a set of objects and their locations. We propose OBJ2TEXT, a sequence-to-sequence model that encodes a set of objects and their locations as an input sequence using an LSTM network, and decodes this representation using an LSTM language model. We show that our model, despite encoding object layouts as a sequence, can represent spatial relationships between objects, and generate descriptions that are globally coherent and semantically relevant. We test our approach in a task of object-layout captioning by using only object annotations as inputs. We additionally show that our model, combined with a state-of-the-art object detector, improves an image captioning model from 0.863 to 0.950 (CIDEr score) in the test benchmark of the standard MS-COCO Captioning task.

arXiv:1707.08052공식 코드 · Papers with Code

Challenges in Data-to-Document Generation

Recent neural models have shown significant progress on the problem of generating short descriptive texts conditioned on a small number of database records. In this work, we suggest a slightly more difficult data-to-text generation task, and investigate how effective current approaches are on this task. In particular, we introduce a new, large-scale corpus of data records paired with descriptive documents, propose a series of extractive evaluation methods for analyzing performance, and obtain baseline results using current neural generation methods. Experiments show that these models produce fluent text, but fail to convincingly approximate human-generated documents. Moreover, even templated baselines exceed the performance of these neural models on some metrics, though copy- and reconstruction-based extensions lead to noticeable improvements.

arXiv:1611.06949공식 코드 · Papers with Code

Dense Captioning with Joint Inference and Visual Context

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

arXiv:1708.02531

Deep Binaries: Encoding Semantic-Rich Cues for Efficient Textual-Visual Cross Retrieval

Cross-modal hashing is usually regarded as an effective technique for large-scale textual-visual cross retrieval, where data from different modalities are mapped into a shared Hamming space for matching. Most of the traditional textual-visual binary encoding methods only consider holistic image representations and fail to model descriptive sentences. This renders existing methods inappropriate to handle the rich semantics of informative cross-modal data for quality textual-visual search tasks. To address the problem of hashing cross-modal data with semantic-rich cues, in this paper, a novel integrated deep architecture is developed to effectively encode the detailed semantics of informative images and long descriptive sentences, named as Textual-Visual Deep Binaries (TVDB). In particular, region-based convolutional networks with long short-term memory units are introduced to fully explore image regional details while semantic cues of sentences are modeled by a text convolutional network. Additionally, we propose a stochastic batch-wise training routine, where high-quality binary codes and deep encoding functions are efficiently optimized in an alternating manner. Experiments are conducted on three multimedia datasets, i.e. Microsoft COCO, IAPR TC-12, and INRIA Web Queries, where the proposed TVDB model significantly outperforms state-of-the-art binary coding methods in the task of cross-modal retrieval.