PyTorchKR Papers

최신 논문

총 194편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1808.08319공식 코드 · Papers with Code

BOP: Benchmark for 6D Object Pose Estimation

We propose a benchmark for 6D pose estimation of a rigid object from a single RGB-D input image. The training data consists of a texture-mapped 3D object model or images of the object in known 6D poses. The benchmark comprises of: i) eight datasets in a unified format that cover different practical scenarios, including two new datasets focusing on varying lighting conditions, ii) an evaluation methodology with a pose-error function that deals with pose ambiguities, iii) a comprehensive evaluation of 15 diverse recent methods that captures the status quo of the field, and iv) an online evaluation system that is open for continuous submission of new results. The evaluation shows that methods based on point-pair features currently perform best, outperforming template matching methods, learning-based methods and methods based on 3D local features. The project website is available at bop.felk.cvut.cz.

arXiv:1808.06194

Fast and Robust Matching for Multimodal Remote Sensing Image Registration

While image registration has been studied in remote sensing community for decades, registering multimodal data [e.g., optical, LiDAR, SAR, and map] remains a challenging problem because of significant nonlinear intensity differences between such data. To address this problem, this paper presents a fast and robust matching framework integrating local descriptors for multimodal registration. In the proposed framework, a local descriptor, such as Histogram of Oriented Gradient (HOG), Local Self Similarity (LSS), or Speeded-Up Robust Feature (SURF), is first extracted at each pixel to form a pixel-wise feature representation of an image. Then we define a similarity measure based on the feature representation in frequency domain using the 3 Dimensional Fast Fourier Transform (3DFFT) technique, followed by a template matching scheme to detect control points between images. In this procedure, we also propose a novel pixel-wise feature representation using orientated gradients of images, which is named channel features of orientated gradients (CFOG). This novel feature is an extension of the pixel-wise HOG descriptors, and outperforms that both in matching performance and computational efficiency. The major advantage of the proposed framework includes: (1) structural similarity representation using the pixel-wise feature description and (2) high computational efficiency due to the use of 3DFFT. Experimental results on different types of multimodal images show the superior matching performance of the proposed framework than the state-of-the-art methods.The proposed matching framework have been used in the software products of a Chinese listed company. The matlab code is available in this manuscript.

arXiv:1807.08666

ASR-free CNN-DTW keyword spotting using multilingual bottleneck features for almost zero-resource languages

We consider multilingual bottleneck features (BNFs) for nearly zero-resource keyword spotting. This forms part of a United Nations effort using keyword spotting to support humanitarian relief programmes in parts of Africa where languages are severely under-resourced. We use 1920 isolated keywords (40 types, 34 minutes) as exemplars for dynamic time warping (DTW) template matching, which is performed on a much larger body of untranscribed speech. These DTW costs are used as targets for a convolutional neural network (CNN) keyword spotter, giving a much faster system than direct DTW. Here we consider how available data from well-resourced languages can improve this CNN-DTW approach. We show that multilingual BNFs trained on ten languages improve the area under the ROC curve of a CNN-DTW system by 10.9% absolute relative to the MFCC baseline. By combining low-resource DTW-based supervision with information from well-resourced languages, CNN-DTW is a competitive option for low-resource keyword spotting.

arXiv:1807.00637

Mammography Dual View Mass Correspondence

Standard breast cancer screening involves the acquisition of two mammography X-ray projections for each breast. Typically, a comparison of both views supports the challenging task of tumor detection and localization. We introduce a deep learning, patch-based Siamese network for lesion matching in dual-view mammography. Our locally-fitted approach generates a joint patch pair representation and comparison with a shared configuration between the two views. We performed a comprehensive set of experiments with the network on standard datasets, among them the large Digital Database for Screening Mammography (DDSM). We analyzed the effect of transfer learning with the network between different types of datasets and compared the network-based matching to using Euclidean distance by template matching. Finally, we evaluated the contribution of the matching network in a full detection pipeline. Experimental results demonstrate the promise of improved detection accuracy using our approach.

arXiv:1806.09374

Fast ASR-free and almost zero-resource keyword spotting using DTW and CNNs for humanitarian monitoring

We use dynamic time warping (DTW) as supervision for training a convolutional neural network (CNN) based keyword spotting system using a small set of spoken isolated keywords. The aim is to allow rapid deployment of a keyword spotting system in a new language to support urgent United Nations (UN) relief programmes in parts of Africa where languages are extremely under-resourced and the development of annotated speech resources is infeasible. First, we use 1920 recorded keywords (40 keyword types, 34 minutes of speech) as exemplars in a DTW-based template matching system and apply it to untranscribed broadcast speech. Then, we use the resulting DTW scores as targets to train a CNN on the same unlabelled speech. In this way we use just 34 minutes of labelled speech, but leverage a large amount of unlabelled data for training. While the resulting CNN keyword spotter cannot match the performance of the DTW-based system, it substantially outperforms a CNN classifier trained only on the keywords, improving the area under the ROC curve from 0.54 to 0.64. Because our CNN system is several orders of magnitude faster at runtime than the DTW system, it represents the most viable keyword spotter on this extremely limited dataset.

arXiv:1310.6066

Skin Segmentation based Elastic Bunch Graph Matching for efficient multiple Face Recognition

This paper is aimed at developing and combining different algorithms for face detection and face recognition to generate an efficient mechanism that can detect and recognize the facial regions of input image. For the detection of face from complex region, skin segmentation isolates the face-like regions in a complex image and following operations of morphology and template matching rejects false matches to extract facial region. For the recognition of the face, the image database is now converted into a database of facial segments. Hence, implementing the technique of Elastic Bunch Graph matching (EBGM) after skin segmentation generates Face Bunch Graphs that acutely represents the features of an individual face enhances the quality of the training set. This increases the matching probability significantly.

arXiv:1405.4308

Coarse-to-Fine Classification via Parametric and Nonparametric Models for Computer-Aided Diagnosis

Classification is one of the core problems in Computer-Aided Diagnosis (CAD), targeting for early cancer detection using 3D medical imaging interpretation. High detection sensitivity with desirably low false positive (FP) rate is critical for a CAD system to be accepted as a valuable or even indispensable tool in radiologists' workflow. Given various spurious imagery noises which cause observation uncertainties, this remains a very challenging task. In this paper, we propose a novel, two-tiered coarse-to-fine (CTF) classification cascade framework to tackle this problem. We first obtain classification-critical data samples (e.g., samples on the decision boundary) extracted from the holistic data distributions using a robust parametric model (e.g., \cite{Raykar08}); then we build a graph-embedding based nonparametric classifier on sampled data, which can more accurately preserve or formulate the complex classification boundary. These two steps can also be considered as effective "sample pruning" and "feature pursuing + $k$NN/template matching", respectively. Our approach is validated comprehensively in colorectal polyp detection and lung nodule detection CAD systems, as the top two deadly cancers, using hospital scale, multi-site clinical datasets. The results show that our method achieves overall better classification/detection performance than existing state-of-the-art algorithms using single-layer classifiers, such as the support vector machine variants \cite{Wang08}, boosting \cite{Slabaugh10}, logistic regression \cite{Ravesteijn10}, relevance vector machine \cite{Raykar08}, $k$-nearest neighbor \cite{Murphy09} or spectral projections on graph \cite{Cai08}.

arXiv:1407.3535

Optimizing Auto-correlation for Fast Target Search in Large Search Space

In remote sensing image-blurring is induced by many sources such as atmospheric scatter, optical aberration, spatial and temporal sensor integration. The natural blurring can be exploited to speed up target search by fast template matching. In this paper, we synthetically induce additional non-uniform blurring to further increase the speed of the matching process. To avoid loss of accuracy, the amount of synthetic blurring is varied spatially over the image according to the underlying content. We extend transitive algorithm for fast template matching by incorporating controlled image blur. To this end we propose an Efficient Group Size (EGS) algorithm which minimizes the number of similarity computations for a particular search image. A larger efficient group size guarantees less computations and more speedup. EGS algorithm is used as a component in our proposed Optimizing auto-correlation (OptA) algorithm. In OptA a search image is iteratively non-uniformly blurred while ensuring no accuracy degradation at any image location. In each iteration efficient group size and overall computations are estimated by using the proposed EGS algorithm. The OptA algorithm stops when the number of computations cannot be further decreased without accuracy degradation. The proposed algorithm is compared with six existing state of the art exhaustive accuracy techniques using correlation coefficient as the similarity measure. Experiments on satellite and aerial image datasets demonstrate the effectiveness of the proposed algorithm.

arXiv:1408.6762

Chatbot for admissions

The communication of potential students with a university department is performed manually and it is a very time consuming procedure. The opportunity to communicate with on a one-to-one basis is highly valued. However with many hundreds of applications each year, one-to-one conversations are not feasible in most cases. The communication will require a member of academic staff to expend several hours to find suitable answers and contact each student. It would be useful to reduce his costs and time. The project aims to reduce the burden on the head of admissions, and potentially other users, by developing a convincing chatbot. A suitable algorithm must be devised to search through the set of data and find a potential answer. The program then replies to the user and provides a relevant web link if the user is not satisfied by the answer. Furthermore a web interface is provided for both users and an administrator. The achievements of the project can be summarised as follows. To prepare the background of the project a literature review was undertaken, together with an investigation of existing tools, and consultation with the head of admissions. The requirements of the system were established and a range of algorithms and tools were investigated, including keyword and template matching. An algorithm that combines keyword matching with string similarity has been developed. A usable system using the proposed algorithm has been implemented. The system was evaluated by keeping logs of questions and answers and by feedback received by potential students that used it.

arXiv:1409.3024

One-Dimensional Vector based Pattern Matching

Template matching is a basic method in image analysis to extract useful information from images. In this paper, we suggest a new method for pattern matching. Our method transform the template image from two dimensional image into one dimensional vector. Also all sub-windows (same size of template) in the reference image will transform into one dimensional vectors. The three similarity measures SAD, SSD, and Euclidean are used to compute the likeness between template and all sub-windows in the reference image to find the best match. The experimental results show the superior performance of the proposed method over the conventional methods on various template of different sizes.

arXiv:1411.4702

Toward a Universal Cortical Algorithm: Examining Hierarchical Temporal Memory in Light of Frontal Cortical Function

A wide range of evidence points toward the existence of a common algorithm underlying the processing of information throughout the cerebral cortex. Several hypothesized features of this cortical algorithm are reviewed, including sparse distributed representation, Bayesian inference, hierarchical organization composed of alternating template matching and pooling layers, temporal slowness and predictive coding. Hierarchical Temporal Memory (HTM) is a family of learning algorithms and corresponding theories of cortical function that embodies these principles. HTM has previously been applied mainly to perceptual tasks typical of posterior cortex. In order to evaluate HTM as a candidate model of cortical function, it is necessary also to investigate its compatibility with the requirements of frontal cortical function. To this end, a variety of models of frontal cortical function are reviewed and integrated, to arrive at the hypothesis that frontal functions including attention, working memory and action selection depend largely upon the same basic algorithms as do posterior functions, with the notable additions of a mechanism for the active maintenance of representations and of multiple cortico-striato-thalamo-cortical loops that allow communication between regions of frontal cortex to be gated in an adaptive manner. Computational models of this system are reviewed. Finally, there is a discussion of how HTM can contribute to the understanding of frontal cortical function, and of what the requirements of frontal cortical function mean for the future development of HTM.

arXiv:1506.03936

A Novel Hybrid Approach for Cephalometric Landmark Detection

Cephalometric analysis has an important role in dentistry and especially in orthodontics as a treatment planning tool to gauge the size and special relationships of the teeth, jaws and cranium. The first step of using such analyses is localizing some important landmarks known as cephalometric landmarks on craniofacial in x-ray image. The past decade has seen a growing interest in automating this process. In this paper, a novel hybrid approach is proposed for automatic detection of cephalometric landmarks. Here, the landmarks are categorized into three main sets according to their anatomical characteristics and usage in well-known cephalometric analyses. Consequently, to have a reliable and accurate detection system, three methods named edge tracing, weighted template matching, and analysis based estimation are designed, each of which is consistent and well-suited for one category. Edge tracing method is suggested to predict those landmarks which are located on edges. Weighted template matching method is well-suited for landmarks located in an obvious and specific structure which can be extracted or searchable in a given x-ray image. The last but not the least method is named analysis based estimation. This method is based on the fact that in cephalometric analyses the relations between landmarks are used and the locations of some landmarks are never used individually. Therefore the third suggested method has a novelty in estimating the desired relations directly. The effectiveness of the proposed approach is compared with the state of the art methods and the results were promising especially in real world applications.

arXiv:1507.02177

Iris Recognition Using Scattering Transform and Textural Features

Iris recognition has drawn a lot of attention since the mid-twentieth century. Among all biometric features, iris is known to possess a rich set of features. Different features have been used to perform iris recognition in the past. In this paper, two powerful sets of features are introduced to be used for iris recognition: scattering transform-based features and textural features. PCA is also applied on the extracted features to reduce the dimensionality of the feature vector while preserving most of the information of its initial value. Minimum distance classifier is used to perform template matching for each new test sample. The proposed scheme is tested on a well-known iris database, and showed promising results with the best accuracy rate of 99.2%.

arXiv:1509.04863

Fast Template Matching by Subsampled Circulant Matrix

Template matching is widely used for many applications in image and signal processing and usually is time-critical. Traditional methods usually focus on how to reduce the search locations by coarse-to-fine strategy or full search combined with pruning strategy. However, the computation cost of those methods is easily dominated by the size of signal N instead of that of template K. This paper proposes a probabilistic and fast matching scheme, which computation costs requires O(N) additions and O(K \log K) multiplications, based on cross-correlation. The nuclear idea is to first downsample signal, which size becomes O(K), and then subsequent operations only involves downsampled signals. The probability of successful match depends on cross-correlation between signal and the template. We show the sufficient condition for successful match and prove that the probability is high for binary signals with K^2/log K >= O(N). The experiments shows this proposed scheme is fast and efficient and supports the theoretical results.

arXiv:1509.08383

Efficient Discriminative Nonorthogonal Binary Subspace with its Application to Visual Tracking

One of the crucial problems in visual tracking is how the object is represented. Conventional appearance-based trackers are using increasingly more complex features in order to be robust. However, complex representations typically not only require more computation for feature extraction, but also make the state inference complicated. We show that with a careful feature selection scheme, extremely simple yet discriminative features can be used for robust object tracking. The central component of the proposed method is a succinct and discriminative representation of the object using discriminative non-orthogonal binary subspace (DNBS) which is spanned by Haar-like features. The DNBS representation inherits the merits of the original NBS in that it efficiently describes the object. It also incorporates the discriminative information to distinguish foreground from background. However, the problem of finding the DNBS bases from an over-complete dictionary is NP-hard. We propose a greedy algorithm called discriminative optimized orthogonal matching pursuit (D-OOMP) to solve this problem. An iterative formulation named iterative D-OOMP is further developed to drastically reduce the redundant computation between iterations and a hierarchical selection strategy is integrated for reducing the search space of features. The proposed DNBS representation is applied to object tracking through SSD-based template matching. We validate the effectiveness of our method through extensive experiments on challenging videos with comparisons against several state-of-the-art trackers and demonstrate its capability to track objects in clutter and moving background.

arXiv:1503.03429

Dense image registration and deformable surface reconstruction in presence of occlusions and minimal texture

Deformable surface tracking from monocular images is well-known to be under-constrained. Occlusions often make the task even more challenging, and can result in failure if the surface is not sufficiently textured. In this work, we explicitly address the problem of 3D reconstruction of poorly textured, occluded surfaces, proposing a framework based on a template-matching approach that scales dense robust features by a relevancy score. Our approach is extensively compared to current methods employing both local feature matching and dense template alignment. We test on standard datasets as well as on a new dataset (that will be made publicly available) of a sparsely textured, occluded surface. Our framework achieves state-of-the-art results for both well and poorly textured, occluded surfaces.

arXiv:1509.03542

Fingerprint Recognition Using Translation Invariant Scattering Network

Fingerprint recognition has drawn a lot of attention during last decades. Different features and algorithms have been used for fingerprint recognition in the past. In this paper, a powerful image representation called scattering transform/network, is used for recognition. Scattering network is a convolutional network where its architecture and filters are predefined wavelet transforms. The first layer of scattering representation is similar to sift descriptors and the higher layers capture higher frequency content of the signal. After extraction of scattering features, their dimensionality is reduced by applying principal component analysis (PCA). At the end, multi-class SVM is used to perform template matching for the recognition task. The proposed scheme is tested on a well-known fingerprint database and has shown promising results with the best accuracy rate of 98\%.

arXiv:1602.01464

Latent-Class Hough Forests for 6 DoF Object Pose Estimation

In this paper we present Latent-Class Hough Forests, a method for object detection and 6 DoF pose estimation in heavily cluttered and occluded scenarios. We adapt a state of the art template matching feature into a scale-invariant patch descriptor and integrate it into a regression forest using a novel template-based split function. We train with positive samples only and we treat class distributions at the leaf nodes as latent variables. During testing we infer by iteratively updating these distributions, providing accurate estimation of background clutter and foreground occlusions and, thus, better detection rate. Furthermore, as a by-product, our Latent-Class Hough Forests can provide accurate occlusion aware segmentation masks, even in the multi-instance scenario. In addition to an existing public dataset, which contains only single-instance sequences with large amounts of clutter, we have collected two, more challenging, datasets for multiple-instance detection containing heavy 2D and 3D clutter as well as foreground occlusions. We provide extensive experiments on the various parameters of the framework such as patch size, number of trees and number of iterations to infer class distributions at test time. We also evaluate the Latent-Class Hough Forests on all datasets where we outperform state of the art methods.

arXiv:1603.06201

A Survey on Object Detection in Optical Remote Sensing Images

Object detection in optical remote sensing images, being a fundamental but challenging problem in the field of aerial and satellite image analysis, plays an important role for a wide range of applications and is receiving significant attention in recent years. While enormous methods exist, a deep review of the literature concerning generic object detection is still lacking. This paper aims to provide a review of the recent progress in this field. Different from several previously published surveys that focus on a specific object class such as building and road, we concentrate on more generic object categories including, but are not limited to, road, building, tree, vehicle, ship, airport, urban-area. Covering about 270 publications we survey 1) template matching-based object detection methods, 2) knowledge-based object detection methods, 3) object-based image analysis (OBIA)-based object detection methods, 4) machine learning-based object detection methods, and 5) five publicly available datasets and three standard evaluation metrics. We also discuss the challenges of current studies and propose two promising research directions, namely deep learning-based feature representation and weakly supervised learning-based geospatial object detection. It is our hope that this survey will be beneficial for the researchers to have better understanding of this research field.