July 28, 2019

2671 words 13 mins read

Paper Group ANR 262

Simulating Structure-from-Motion. Effect of Super Resolution on High Dimensional Features for Unsupervised Face Recognition in the Wild. Face Alignment with Cascaded Semi-Parametric Deep Greedy Neural Forests. Neural Networks for Text Correction and Completion in Keyboard Decoding. Biometrics-as-a-Service: A Framework to Promote Innovative Biometri …

Simulating Structure-from-Motion


Title	Simulating Structure-from-Motion
Authors	Martin Hahner, Orestis Varesis, Panagiotis Bountouris
Abstract	The implementation of a Structure-from-Motion (SfM) pipeline from a synthetically generated scene as well as the investigation of the faithfulness of diverse reconstructions is the subject of this project. A series of different SfM reconstructions are implemented and their camera pose estimations are being contrasted with their respective ground truth locations. Finally, injection of ground truth location data into the rendered images in order to reduce the estimation error of the camera poses is studied as well.
Tasks
Published	2017-10-03
URL	http://arxiv.org/abs/1710.01052v1
PDF	http://arxiv.org/pdf/1710.01052v1.pdf
PWC	https://paperswithcode.com/paper/simulating-structure-from-motion
Repo
Framework

Effect of Super Resolution on High Dimensional Features for Unsupervised Face Recognition in the Wild


Title	Effect of Super Resolution on High Dimensional Features for Unsupervised Face Recognition in the Wild
Authors	Ahmed ElSayed, Ausif Mahmood, Tarek Sobh
Abstract	Majority of the face recognition algorithms use query faces captured from uncontrolled, in the wild, environment. Often caused by the cameras limited capabilities, it is common for these captured facial images to be blurred or low resolution. Super resolution algorithms are therefore crucial in improving the resolution of such images especially when the image size is small requiring enlargement. This paper aims to demonstrate the effect of one of the state-of-the-art algorithms in the field of image super resolution. To demonstrate the functionality of the algorithm, various before and after 3D face alignment cases are provided using the images from the Labeled Faces in the Wild (lfw). Resulting images are subject to testing on a closed set face recognition protocol using unsupervised algorithms with high dimension extracted features. The inclusion of super resolution algorithm resulted in significant improved recognition rate over recently reported results obtained from unsupervised algorithms.
Tasks	Face Alignment, Face Recognition, Image Super-Resolution, Super-Resolution
Published	2017-03-23
URL	http://arxiv.org/abs/1704.01464v2
PDF	http://arxiv.org/pdf/1704.01464v2.pdf
PWC	https://paperswithcode.com/paper/effect-of-super-resolution-on-high
Repo
Framework

Face Alignment with Cascaded Semi-Parametric Deep Greedy Neural Forests


Title	Face Alignment with Cascaded Semi-Parametric Deep Greedy Neural Forests
Authors	Arnaud Dapogny, Kévin Bailly, Séverine Dubuisson
Abstract	Face alignment is an active topic in computer vision, consisting in aligning a shape model on the face. To this end, most modern approaches refine the shape in a cascaded manner, starting from an initial guess. Those shape updates can either be applied in the feature point space (\textit{i.e.} explicit updates) or in a low-dimensional, parametric space. In this paper, we propose a semi-parametric cascade that first aligns a parametric shape, then captures more fine-grained deformations of an explicit shape. For the purpose of learning shape updates at each cascade stage, we introduce a deep greedy neural forest (GNF) model, which is an improved version of deep neural forest (NF). GNF appears as an ideal regressor for face alignment, as it combines differentiability, high expressivity and fast evaluation runtime. The proposed framework is very fast and achieves high accuracies on multiple challenging benchmarks, including small, medium and large pose experiments.
Tasks	Face Alignment
Published	2017-03-05
URL	http://arxiv.org/abs/1703.01597v1
PDF	http://arxiv.org/pdf/1703.01597v1.pdf
PWC	https://paperswithcode.com/paper/face-alignment-with-cascaded-semi-parametric
Repo
Framework

Neural Networks for Text Correction and Completion in Keyboard Decoding


Title	Neural Networks for Text Correction and Completion in Keyboard Decoding
Authors	Shaona Ghosh, Per Ola Kristensson
Abstract	Despite the ubiquity of mobile and wearable text messaging applications, the problem of keyboard text decoding is not tackled sufficiently in the light of the enormous success of the deep learning Recurrent Neural Network (RNN) and Convolutional Neural Networks (CNN) for natural language understanding. In particular, considering that the keyboard decoders should operate on devices with memory and processor resource constraints, makes it challenging to deploy industrial scale deep neural network (DNN) models. This paper proposes a sequence-to-sequence neural attention network system for automatic text correction and completion. Given an erroneous sequence, our model encodes character level hidden representations and then decodes the revised sequence thus enabling auto-correction and completion. We achieve this by a combination of character level CNN and gated recurrent unit (GRU) encoder along with and a word level gated recurrent unit (GRU) attention decoder. Unlike traditional language models that learn from billions of words, our corpus size is only 12 million words; an order of magnitude smaller. The memory footprint of our learnt model for inference and prediction is also an order of magnitude smaller than the conventional language model based text decoders. We report baseline performance for neural keyboard decoders in such limited domain. Our models achieve a word level accuracy of $90%$ and a character error rate CER of $2.4%$ over the Twitter typo dataset. We present a novel dataset of noisy to corrected mappings by inducing the noise distribution from the Twitter data over the OpenSubtitles 2009 dataset; on which our model predicts with a word level accuracy of $98%$ and sequence accuracy of $68.9%$. In our user study, our model achieved an average CER of $2.6%$ with the state-of-the-art non-neural touch-screen keyboard decoder at CER of $1.6%$.
Tasks	Language Modelling
Published	2017-09-19
URL	http://arxiv.org/abs/1709.06429v1
PDF	http://arxiv.org/pdf/1709.06429v1.pdf
PWC	https://paperswithcode.com/paper/neural-networks-for-text-correction-and
Repo
Framework

Biometrics-as-a-Service: A Framework to Promote Innovative Biometric Recognition in the Cloud


Title	Biometrics-as-a-Service: A Framework to Promote Innovative Biometric Recognition in the Cloud
Authors	Veeru Talreja, Terry Ferrett, Matthew C. Valenti, Arun Ross
Abstract	Biometric recognition, or simply biometrics, is the use of biological attributes such as face, fingerprints or iris in order to recognize an individual in an automated manner. A key application of biometrics is authentication; i.e., using said biological attributes to provide access by verifying the claimed identity of an individual. This paper presents a framework for Biometrics-as-a-Service (BaaS) that performs biometric matching operations in the cloud, while relying on simple and ubiquitous consumer devices such as smartphones. Further, the framework promotes innovation by providing interfaces for a plurality of software developers to upload their matching algorithms to the cloud. When a biometric authentication request is submitted, the system uses a criteria to automatically select an appropriate matching algorithm. Every time a particular algorithm is selected, the corresponding developer is rendered a micropayment. This creates an innovative and competitive ecosystem that benefits both software developers and the consumers. As a case study, we have implemented the following: (a) an ocular recognition system using a mobile web interface providing user access to a biometric authentication service, and (b) a Linux-based virtual machine environment used by software developers for algorithm development and submission.
Tasks
Published	2017-10-25
URL	http://arxiv.org/abs/1710.09183v1
PDF	http://arxiv.org/pdf/1710.09183v1.pdf
PWC	https://paperswithcode.com/paper/biometrics-as-a-service-a-framework-to
Repo
Framework

Image Augmentation using Radial Transform for Training Deep Neural Networks


Title	Image Augmentation using Radial Transform for Training Deep Neural Networks
Authors	Hojjat Salehinejad, Shahrokh Valaee, Timothy Dowdell, Joseph Barfett
Abstract	Deep learning models have a large number of free parameters that must be estimated by efficient training of the models on a large number of training data samples to increase their generalization performance. In real-world applications, the data available to train these networks is often limited or imbalanced. We propose a sampling method based on the radial transform in a polar coordinate system for image augmentation to facilitate the training of deep learning models from limited source data. This pixel-wise transform provides representations of the original image in the polar coordinate system by generating a new image from each pixel. This technique can generate radial transformed images up to the number of pixels in the original image to increase the diversity of poorly represented image classes. Our experiments show improved generalization performance in training deep convolutional neural networks with radial transformed images.
Tasks	Image Augmentation
Published	2017-08-14
URL	http://arxiv.org/abs/1708.04347v4
PDF	http://arxiv.org/pdf/1708.04347v4.pdf
PWC	https://paperswithcode.com/paper/image-augmentation-using-radial-transform-for
Repo
Framework

A step towards procedural terrain generation with GANs


Title	A step towards procedural terrain generation with GANs
Authors	Christopher Beckham, Christopher Pal
Abstract	Procedural terrain generation for video games has been traditionally been done with smartly designed but handcrafted algorithms that generate heightmaps. We propose a first step toward the learning and synthesis of these using recent advances in deep generative modelling with openly available satellite imagery from NASA.
Tasks
Published	2017-07-11
URL	http://arxiv.org/abs/1707.03383v1
PDF	http://arxiv.org/pdf/1707.03383v1.pdf
PWC	https://paperswithcode.com/paper/a-step-towards-procedural-terrain-generation
Repo
Framework

Compressing Green’s function using intermediate representation between imaginary-time and real-frequency domains


Title	Compressing Green’s function using intermediate representation between imaginary-time and real-frequency domains
Authors	Hiroshi Shinaoka, Junya Otsuki, Masayuki Ohzeki, Kazuyoshi Yoshimi
Abstract	New model-independent compact representations of imaginary-time data are presented in terms of the intermediate representation (IR) of analytical continuation. This is motivated by a recent numerical finding by the authors [J. Otsuki et al., arXiv:1702.03056]. We demonstrate the efficiency of the IR through continuous-time quantum Monte Carlo calculations of an Anderson impurity model. We find that the IR yields a significantly compact form of various types of correlation functions. The present framework will provide general ways to boost the power of cutting-edge diagrammatic/quantum Monte Carlo treatments of many-body systems.
Tasks
Published	2017-02-10
URL	http://arxiv.org/abs/1702.03054v3
PDF	http://arxiv.org/pdf/1702.03054v3.pdf
PWC	https://paperswithcode.com/paper/compressing-greens-function-using
Repo
Framework

Identity and Granularity of Events in Text


Title	Identity and Granularity of Events in Text
Authors	Piek Vossen, Agata Cybulska
Abstract	In this paper we describe a method to detect event descrip- tions in different news articles and to model the semantics of events and their components using RDF representations. We compare these descriptions to solve a cross-document event coreference task. Our com- ponent approach to event semantics defines identity and granularity of events at different levels. It performs close to state-of-the-art approaches on the cross-document event coreference task, while outperforming other works when assuming similar quality of event detection. We demonstrate how granularity and identity are interconnected and we discuss how se- mantic anomaly could be used to define differences between coreference, subevent and topical relations.
Tasks
Published	2017-04-13
URL	http://arxiv.org/abs/1704.04259v1
PDF	http://arxiv.org/pdf/1704.04259v1.pdf
PWC	https://paperswithcode.com/paper/identity-and-granularity-of-events-in-text
Repo
Framework

The WILDTRACK Multi-Camera Person Dataset


Title	The WILDTRACK Multi-Camera Person Dataset
Authors	Tatjana Chavdarova, Pierre Baqué, Stéphane Bouquet, Andrii Maksai, Cijo Jose, Louis Lettry, Pascal Fua, Luc Van Gool, François Fleuret
Abstract	People detection methods are highly sensitive to the perpetual occlusions among the targets. As multi-camera set-ups become more frequently encountered, joint exploitation of the across views information would allow for improved detection performances. We provide a large-scale HD dataset named WILDTRACK which finally makes advanced deep learning methods applicable to this problem. The seven-static-camera set-up captures realistic and challenging scenarios of walking people. Notably, its camera calibration with jointly high-precision projection widens the range of algorithms which may make use of this dataset. In aim to help accelerate the research on automatic camera calibration, such annotations also accompany this dataset. Furthermore, the rich-in-appearance visual context of the pedestrian class makes this dataset attractive for monocular pedestrian detection as well, since: the HD cameras are placed relatively close to the people, and the size of the dataset further increases seven-fold. In summary, we overview existing multi-camera datasets and detection methods, enumerate details of our dataset, and we benchmark multi-camera state of the art detectors on this new dataset.
Tasks	Calibration, Pedestrian Detection
Published	2017-07-28
URL	http://arxiv.org/abs/1707.09299v1
PDF	http://arxiv.org/pdf/1707.09299v1.pdf
PWC	https://paperswithcode.com/paper/the-wildtrack-multi-camera-person-dataset
Repo
Framework

Automatic Streaming Segmentation of Stereo Video Using Bilateral Space


Title	Automatic Streaming Segmentation of Stereo Video Using Bilateral Space
Authors	Wenjing Ke, Yuanjie Zhu, Lei Yu
Abstract	In this paper, we take advantage of binocular camera and propose an unsupervised algorithm based on semi-supervised segmentation algorithm and extracting foreground part efficiently. We creatively embed depth information into bilateral grid in the graph cut model and achieve considerable segmenting accuracy in the case of no user input. The experi- ment approves the high precision, time efficiency of our algorithm and its adaptation to complex natural scenario which is significant for practical application.
Tasks
Published	2017-10-10
URL	http://arxiv.org/abs/1710.03488v2
PDF	http://arxiv.org/pdf/1710.03488v2.pdf
PWC	https://paperswithcode.com/paper/automatic-streaming-segmentation-of-stereo
Repo
Framework

A Proximity-Aware Hierarchical Clustering of Faces


Title	A Proximity-Aware Hierarchical Clustering of Faces
Authors	Wei-An Lin, Jun-Cheng Chen, Rama Chellappa
Abstract	In this paper, we propose an unsupervised face clustering algorithm called “Proximity-Aware Hierarchical Clustering” (PAHC) that exploits the local structure of deep representations. In the proposed method, a similarity measure between deep features is computed by evaluating linear SVM margins. SVMs are trained using nearest neighbors of sample data, and thus do not require any external training data. Clusters are then formed by thresholding the similarity scores. We evaluate the clustering performance using three challenging unconstrained face datasets, including Celebrity in Frontal-Profile (CFP), IARPA JANUS Benchmark A (IJB-A), and JANUS Challenge Set 3 (JANUS CS3) datasets. Experimental results demonstrate that the proposed approach can achieve significant improvements over state-of-the-art methods. Moreover, we also show that the proposed clustering algorithm can be applied to curate a set of large-scale and noisy training dataset while maintaining sufficient amount of images and their variations due to nuisance factors. The face verification performance on JANUS CS3 improves significantly by finetuning a DCNN model with the curated MS-Celeb-1M dataset which contains over three million face images.
Tasks	Face Verification
Published	2017-03-14
URL	http://arxiv.org/abs/1703.04835v1
PDF	http://arxiv.org/pdf/1703.04835v1.pdf
PWC	https://paperswithcode.com/paper/a-proximity-aware-hierarchical-clustering-of
Repo
Framework

Building a Sentiment Corpus of Tweets in Brazilian Portuguese


Title	Building a Sentiment Corpus of Tweets in Brazilian Portuguese
Authors	Henrico Bertini Brum, Maria das Graças Volpe Nunes
Abstract	The large amount of data available in social media, forums and websites motivates researches in several areas of Natural Language Processing, such as sentiment analysis. The popularity of the area due to its subjective and semantic characteristics motivates research on novel methods and approaches for classification. Hence, there is a high demand for datasets on different domains and different languages. This paper introduces TweetSentBR, a sentiment corpora for Brazilian Portuguese manually annotated with 15.000 sentences on TV show domain. The sentences were labeled in three classes (positive, neutral and negative) by seven annotators, following literature guidelines for ensuring reliability on the annotation. We also ran baseline experiments on polarity classification using three machine learning methods, reaching 80.99% on F-Measure and 82.06% on accuracy in binary classification, and 59.85% F-Measure and 64.62% on accuracy on three point classification.
Tasks	Sentiment Analysis
Published	2017-12-24
URL	http://arxiv.org/abs/1712.08917v1
PDF	http://arxiv.org/pdf/1712.08917v1.pdf
PWC	https://paperswithcode.com/paper/building-a-sentiment-corpus-of-tweets-in-1
Repo
Framework

Discriminative k-shot learning using probabilistic models


Title	Discriminative k-shot learning using probabilistic models
Authors	Matthias Bauer, Mateo Rojas-Carulla, Jakub Bartłomiej Świątkowski, Bernhard Schölkopf, Richard E. Turner
Abstract	This paper introduces a probabilistic framework for k-shot image classification. The goal is to generalise from an initial large-scale classification task to a separate task comprising new classes and small numbers of examples. The new approach not only leverages the feature-based representation learned by a neural network from the initial task (representational transfer), but also information about the classes (concept transfer). The concept information is encapsulated in a probabilistic model for the final layer weights of the neural network which acts as a prior for probabilistic k-shot learning. We show that even a simple probabilistic model achieves state-of-the-art on a standard k-shot learning dataset by a large margin. Moreover, it is able to accurately model uncertainty, leading to well calibrated classifiers, and is easily extensible and flexible, unlike many recent approaches to k-shot learning.
Tasks	Few-Shot Image Classification, Image Classification
Published	2017-06-01
URL	http://arxiv.org/abs/1706.00326v2
PDF	http://arxiv.org/pdf/1706.00326v2.pdf
PWC	https://paperswithcode.com/paper/discriminative-k-shot-learning-using
Repo
Framework

Imagining Probabilistic Belief Change as Imaging (Technical Report)


Title	Imagining Probabilistic Belief Change as Imaging (Technical Report)
Authors	Gavin Rens, Thomas Meyer
Abstract	Imaging is a form of probabilistic belief change which could be employed for both revision and update. In this paper, we propose a new framework for probabilistic belief change based on imaging, called Expected Distance Imaging (EDI). EDI is sufficiently general to define Bayesian conditioning and other forms of imaging previously defined in the literature. We argue that, and investigate how, EDI can be used for both revision and update. EDI’s definition depends crucially on a weight function whose properties are studied and whose effect on belief change operations is analysed. Finally, four EDI instantiations are proposed, two for revision and two for update, and probabilistic rationality postulates are suggested for their analysis.
Tasks
Published	2017-05-02
URL	http://arxiv.org/abs/1705.01172v1
PDF	http://arxiv.org/pdf/1705.01172v1.pdf
PWC	https://paperswithcode.com/paper/imagining-probabilistic-belief-change-as
Repo
Framework