Home [81–90] << 91 92 93 94 95 96 97 98 99 100 >> [101–110]
List View
 | 
Citations
 | 
Details
   web
Records
Author Andres Mafla
Title (down) Leveraging Scene Text Information for Image Interpretation Type Book Whole
Year 2022 Publication PhD Thesis, Universitat Autonoma de Barcelona-CVC Abbreviated Journal
Volume Issue Pages
Keywords
Abstract Until recently, most computer vision models remained illiterate, largely ignoring the semantically rich and explicit information contained in scene text. Recent progress in scene text detection and recognition has recently allowed exploring its role in a diverse set of open computer vision problems, e.g. image classification, image-text retrieval, image captioning, and visual question answering to name a few. The explicit semantics of scene text closely requires specific modeling similar to language. However, scene text is a particular signal that has to be interpreted according to a comprehensive perspective that encapsulates all the visual cues in an image. Incorporating this information is a straightforward task for humans, but if we are unfamiliar with a language or scripture, achieving a complete world understanding is impossible (e.a. visiting a foreign country with a different alphabet). Despite the importance of scene text, modeling it requires considering the several ways in which scene text interacts with an image, processing and fusing an additional modality. In this thesis, we mainly focus
on two tasks, scene text-based fine-grained image classification, and cross-modal retrieval. In both studied tasks we identify existing limitations in current approaches and propose plausible solutions. Concretely, in each chapter: i) We define a compact way to embed scene text that generalizes to unseen words at training time while performing in real-time. ii) We incorporate the previously learned scene text embedding to create an image-level descriptor that overcomes optical character recognition (OCR) errors which is well-suited to the fine-grained image classification task. iii) We design a region-level reasoning network that learns the interaction through semantics among salient visual regions and scene text instances. iv) We employ scene text information in image-text matching and introduce the Scene Text Aware Cross-Modal retrieval StacMR task. We gather a dataset that incorporates scene text and design a model suited for the newly studied modality. v) We identify the drawbacks of current retrieval metrics in cross-modal retrieval. An image captioning metric is proposed as a way of better evaluating semantics in retrieved results. Ample experimentation shows that incorporating such semantics into a model yields better semantic results while
requiring significantly less data to converge.
Address
Corporate Author Thesis Ph.D. thesis
Publisher IMPRIMA Place of Publication Editor Dimosthenis Karatzas;Lluis Gomez
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN 978-84-124793-6-2 Medium
Area Expedition Conference
Notes DAG Approved no
Call Number Admin @ si @ Maf2022 Serial 3756
Permanent link to this record
 
 
Author Svebor Karaman; Giuseppe Lisanti; Andrew Bagdanov; Alberto del Bimbo
Title (down) Leveraging local neighborhood topology for large scale person re-identification Type Journal Article
Year 2014 Publication Pattern Recognition Abbreviated Journal PR
Volume 47 Issue 12 Pages 3767–3778
Keywords Re-identification; Conditional random field; Semi-supervised; ETHZ; CAVIAR; 3DPeS; CMV100
Abstract In this paper we describe a semi-supervised approach to person re-identification that combines discriminative models of person identity with a Conditional Random Field (CRF) to exploit the local manifold approximation induced by the nearest neighbor graph in feature space. The linear discriminative models learned on few gallery images provides coarse separation of probe images into identities, while a graph topology defined by distances between all person images in feature space leverages local support for label propagation in the CRF. We evaluate our approach using multiple scenarios on several publicly available datasets, where the number of identities varies from 28 to 191 and the number of images ranges between 1003 and 36 171. We demonstrate that the discriminative model and the CRF are complementary and that the combination of both leads to significant improvement over state-of-the-art approaches. We further demonstrate how the performance of our approach improves with increasing test data and also with increasing amounts of additional unlabeled data.
Address
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference
Notes LAMP; 601.240; 600.079 Approved no
Call Number Admin @ si @ KLB2014a Serial 2522
Permanent link to this record
 
 
Author Albert Gordo; Jose Antonio Rodriguez; Florent Perronnin; Ernest Valveny
Title (down) Leveraging category-level labels for instance-level image retrieval Type Conference Article
Year 2012 Publication 25th IEEE Conference on Computer Vision and Pattern Recognition Abbreviated Journal
Volume Issue Pages 3045-3052
Keywords
Abstract In this article, we focus on the problem of large-scale instance-level image retrieval. For efficiency reasons, it is common to represent an image by a fixed-length descriptor which is subsequently encoded into a small number of bits. We note that most encoding techniques include an unsupervised dimensionality reduction step. Our goal in this work is to learn a better subspace in a supervised manner. We especially raise the following question: “can category-level labels be used to learn such a subspace?” To answer this question, we experiment with four learning techniques: the first one is based on a metric learning framework, the second one on attribute representations, the third one on Canonical Correlation Analysis (CCA) and the fourth one on Joint Subspace and Classifier Learning (JSCL). While the first three approaches have been applied in the past to the image retrieval problem, we believe we are the first to show the usefulness of JSCL in this context. In our experiments, we use ImageNet as a source of category-level labels and report retrieval results on two standard dataseis: INRIA Holidays and the University of Kentucky benchmark. Our experimental study shows that metric learning and attributes do not lead to any significant improvement in retrieval accuracy, as opposed to CCA and JSCL. As an example, we report on Holidays an increase in accuracy from 39.3% to 48.6% with 32-dimensional representations. Overall JSCL is shown to yield the best results.
Address Providence, Rhode Island
Corporate Author Thesis
Publisher IEEE Xplore Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN 1063-6919 ISBN 978-1-4673-1226-4 Medium
Area Expedition Conference CVPR
Notes DAG Approved no
Call Number Admin @ si @ GRP2012 Serial 2050
Permanent link to this record
 
 
Author Ali Furkan Biten; Lluis Gomez; Dimosthenis Karatzas
Title (down) Let there be a clock on the beach: Reducing Object Hallucination in Image Captioning Type Conference Article
Year 2022 Publication Winter Conference on Applications of Computer Vision Abbreviated Journal
Volume Issue Pages 1381-1390
Keywords Measurement; Training; Visualization; Analytical models; Computer vision; Computational modeling; Training data
Abstract Explaining an image with missing or non-existent objects is known as object bias (hallucination) in image captioning. This behaviour is quite common in the state-of-the-art captioning models which is not desirable by humans. To decrease the object hallucination in captioning, we propose three simple yet efficient training augmentation method for sentences which requires no new training data or increase
in the model size. By extensive analysis, we show that the proposed methods can significantly diminish our models’ object bias on hallucination metrics. Moreover, we experimentally demonstrate that our methods decrease the dependency on the visual features. All of our code, configuration files and model weights are available online.
Address Virtual; Waikoloa; Hawai; USA; January 2022
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference WACV
Notes DAG; 600.155; 302.105 Approved no
Call Number Admin @ si @ BGK2022 Serial 3662
Permanent link to this record
 
 
Author G. de Oliveira; A. Cartas; Marc Bolaños; Mariella Dimiccoli; Xavier Giro; Petia Radeva
Title (down) LEMoRe: A Lifelog Engine for Moments Retrieval at the NTCIR-Lifelog LSAT Task Type Conference Article
Year 2016 Publication 12th NTCIR Conference on Evaluation of Information Access Technologies Abbreviated Journal
Volume Issue Pages
Keywords
Abstract Semantic image retrieval from large amounts of egocentric visual data requires to leverage powerful techniques for filling in the semantic gap. This paper introduces LEMoRe, a Lifelog Engine for Moments Retrieval, developed in the context of the Lifelog Semantic Access Task (LSAT) of the the NTCIR-12 challenge and discusses its performance variation on different trials. LEMoRe integrates classical image descriptors with high-level semantic concepts extracted by Convolutional Neural Networks (CNN), powered by a graphic user interface that uses natural language processing. Although this is just a first attempt towards interactive image retrieval from large egocentric datasets and there is a large room for improvement of the system components and the user interface, the structure of the system itself and the way the single components cooperate are very promising.
Address Tokyo; Japan; June 2016
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference NTCIR
Notes MILAB; Approved no
Call Number Admin @ si @OCB2016 Serial 2789
Permanent link to this record
 
 
Author C. Butakoff; Simone Balocco; F.M. Sukno; C. Hoogendoorn; C. Tobon-Gomez; G. Avegliano; A.F. Frangi
Title (down) Left-ventricular Epi- and Endocardium Extraction from 3D Ultrasound Images Using an Automatically Constructed 3D ASM Type Journal Article
Year 2016 Publication Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization Abbreviated Journal CMBBE
Volume 4 Issue 5 Pages 265-280
Keywords ASM; cardiac segmentation; statistical model; shape model; 3D ultrasound; cardiac segmentation
Abstract In this paper, we propose an automatic method for constructing an active shape model (ASM) to segment the complete cardiac left ventricle in 3D ultrasound (3DUS) images, which avoids costly manual landmarking. The automatic construction of the ASM has already been addressed in the literature; however, the direct application of these methods to 3DUS is hampered by a high level of noise and artefacts. Therefore, we propose to construct the ASM by fusing the multidetector computed tomography data, to learn the shape, with the artificially generated 3DUS, in order to learn the neighbourhood of the boundaries. Our artificial images were generated by two approaches: a faster one that does not take into account the geometry of the transducer, and a more comprehensive one, implemented in Field II toolbox. The segmentation accuracy of our ASM was evaluated on 20 patients with left-ventricular asynchrony, demonstrating plausibility of the approach.
Address
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN 2168-1163 ISBN Medium
Area Expedition Conference
Notes MILAB Approved no
Call Number Admin @ si @ BBS2016 Serial 2449
Permanent link to this record
 
 
Author Antoni Gurgui; Debora Gil; Enric Marti; Vicente Grau
Title (down) Left-Ventricle Basal Region Constrained Parametric Mapping to Unitary Domain Type Conference Article
Year 2016 Publication 7th International Workshop on Statistical Atlases & Computational Modelling of the Heart Abbreviated Journal
Volume 10124 Issue Pages 163-171
Keywords Laplacian; Constrained maps; Parameterization; Basal ring
Abstract Due to its complex geometry, the basal ring is often omitted when putting different heart geometries into correspondence. In this paper, we present the first results on a new mapping of the left ventricle basal rings onto a normalized coordinate system using a fold-over free approach to the solution to the Laplacian. To guarantee correspondences between different basal rings, we imposed some internal constrained positions at anatomical landmarks in the normalized coordinate system. To prevent internal fold-overs, constraints are handled by cutting the volume into regions defined by anatomical features and mapping each piece of the volume separately. Initial results presented in this paper indicate that our method is able to handle internal constrains without introducing fold-overs and thus guarantees one-to-one mappings between different basal ring geometries.
Address Athens; October 2016
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title LNCS
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference STACOM
Notes IAM; Approved no
Call Number Admin @ si @ GGM2016 Serial 2884
Permanent link to this record
 
 
Author Francesc Carreras; Jaume Garcia; Debora Gil; Sandra Pujadas; Chi ho Lion; R.Suarez-Arias; R.Leta; Xavier Alomar; Manuel Ballester; Guillem Pons-Llados
Title (down) Left ventricular torsion and longitudinal shortening: two fundamental components of myocardial mechanics assessed by tagged cine-MRI in normal subjects Type Journal Article
Year 2012 Publication International Journal of Cardiovascular Imaging Abbreviated Journal IJCI
Volume 28 Issue 2 Pages 273-284
Keywords Magnetic resonance imaging (MRI); Tagging MRI; Cardiac mechanics; Ventricular torsion
Abstract Cardiac magnetic resonance imaging (Cardiac MRI) has become a gold standard diagnostic technique for the assessment of cardiac mechanics, allowing the non-invasive calculation of left ventric- ular long axis longitudinal shortening (LVLS) and absolute myocardial torsion (AMT) between basal and apical left ventricular slices, a movement directly related to the helicoidal anatomic disposition of the myocardial fibers. The aim of this study is to determine AMT and LVLS behaviour and normal values from a group of healthy subjects. A group of 21 healthy volunteers (15 males) (age: 23–55 y.o., mean:30.7 ± 7.5) were prospectively included in an obser- vational study by Cardiac MRI. Left ventricular rotation (degrees) was calculated by custom-made software (Harmonic Phase Flow) in consecutive LV short axis planes tagged cine-MRI sequences. AMT was determined from the difference between basal and apical planes LV rotations. LVLS (%) was determined from the LV longitudinal and horizontal axis cine-MRI images. All the 21 cases studied were interpretable, although in three cases the value of the LV apical rotation could not be determined. The mean rotation of the basal and apical planes at end-systole were -3.71° ± 0.84° and 6.73° ± 1.69° (n:18) respectively, resulting in a LV mean AMT of 10.48° ± 1.63° (n:18). End-systolic mean LVLS was 19.07 ± 2.71%. Cardiac MRI allows for the calculation of AMT and LVLS, fundamental functional components of the ventricular twist mechanics conditioned, in turn, by the anatomical helical layout of the myocardial fibers. These values provide complementary information about systolic ventricular function in relation to the traditional parameters used in daily practice.
Address
Corporate Author Thesis
Publisher Springer Netherlands Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN 1569-5794 ISBN Medium
Area Expedition Conference
Notes IAM; Approved no
Call Number IAM @ iam @ CGG2012 Serial 1496
Permanent link to this record
 
 
Author Saad Minhas; Aura Hernandez-Sabate; Shoaib Ehsan; Katerine Diaz; Ales Leonardis; Antonio Lopez; Klaus McDonald Maier
Title (down) LEE: A photorealistic Virtual Environment for Assessing Driver-Vehicle Interactions in Self-Driving Mode Type Conference Article
Year 2016 Publication 14th European Conference on Computer Vision Workshops Abbreviated Journal
Volume 9915 Issue Pages 894-900
Keywords Simulation environment; Automated Driving; Driver-Vehicle interaction
Abstract Photorealistic virtual environments are crucial for developing and testing automated driving systems in a safe way during trials. As commercially available simulators are expensive and bulky, this paper presents a low-cost, extendable, and easy-to-use (LEE) virtual environment with the aim to highlight its utility for level 3 driving automation. In particular, an experiment is performed using the presented simulator to explore the influence of different variables regarding control transfer of the car after the system was driving autonomously in a highway scenario. The results show that the speed of the car at the time when the system needs to transfer the control to the human driver is critical.
Address Amsterdam; The Netherlands; October 2016
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title LNCS
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference ECCVW
Notes ADAS;IAM; 600.085; 600.076 Approved no
Call Number MHE2016 Serial 2865
Permanent link to this record
 
 
Author Eloi Puertas; Miguel Angel Bautista; Daniel Sanchez; Sergio Escalera; Oriol Pujol
Title (down) Learning to Segment Humans by Stacking their Body Parts, Type Conference Article
Year 2014 Publication ECCV Workshop on ChaLearn Looking at People Abbreviated Journal
Volume 8925 Issue Pages 685-697
Keywords Human body segmentation; Stacked Sequential Learning
Abstract Human segmentation in still images is a complex task due to the wide range of body poses and drastic changes in environmental conditions. Usually, human body segmentation is treated in a two-stage fashion. First, a human body part detection step is performed, and then, human part detections are used as prior knowledge to be optimized by segmentation strategies. In this paper, we present a two-stage scheme based on Multi-Scale Stacked Sequential Learning (MSSL). We define an extended feature set by stacking a multi-scale decomposition of body
part likelihood maps. These likelihood maps are obtained in a first stage
by means of a ECOC ensemble of soft body part detectors. In a second stage, contextual relations of part predictions are learnt by a binary classifier, obtaining an accurate body confidence map. The obtained confidence map is fed to a graph cut optimization procedure to obtain the final segmentation. Results show improved segmentation when MSSL is included in the human segmentation pipeline.
Address
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title LNCS
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference ECCVW
Notes HuPBA;MILAB Approved no
Call Number Admin @ si @ PBS2014 Serial 2553
Permanent link to this record
 
 
Author Jon Almazan
Title (down) Learning to Represent Handwritten Shapes and Words for Matching and Recognition Type Book Whole
Year 2014 Publication PhD Thesis, Universitat Autonoma de Barcelona-CVC Abbreviated Journal
Volume Issue Pages
Keywords
Abstract Writing is one of the most important forms of communication and for centuries, handwriting had been the most reliable way to preserve knowledge. However, despite the recent development of printing houses and electronic devices, handwriting is still broadly used for taking notes, doing annotations, or sketching ideas.
Transferring the ability of understanding handwritten text or recognizing handwritten shapes to computers has been the goal of many researches due to its huge importance for many different fields. However, designing good representations to deal with handwritten shapes, e.g. symbols or words, is a very challenging problem due to the large variability of these kinds of shapes. One of the consequences of working with handwritten shapes is that we need representations to be robust, i.e., able to adapt to large intra-class variability. We need representations to be discriminative, i.e., able to learn what are the differences between classes. And, we need representations to be efficient, i.e., able to be rapidly computed and compared. Unfortunately, current techniques of handwritten shape representation for matching and recognition do not fulfill some or all of these requirements.
Through this thesis we focus on the problem of learning to represent handwritten shapes aimed at retrieval and recognition tasks. Concretely, on the first part of the thesis, we focus on the general problem of representing any kind of handwritten shape. We first present a novel shape descriptor based on a deformable grid that deals with large deformations by adapting to the shape and where the cells of the grid can be used to extract different features. Then, we propose to use this descriptor to learn statistical models, based on the Active Appearance Model, that jointly learns the variability in structure and texture of a given class. Then, on the second part, we focus on a concrete application, the problem of representing handwritten words, for the tasks of word spotting, where the goal is to find all instances of a query word in a dataset of images, and recognition. First, we address the segmentation-free problem and propose an unsupervised, sliding-window-based approach that achieves state-of- the-art results in two public datasets. Second, we address the more challenging multi-writer problem, where the variability in words exponentially increases. We describe an approach in which both word images and text strings are embedded in a common vectorial subspace, and where those that represent the same word are close together. This is achieved by a combination of label embedding and attributes learning, and a common subspace regression. This leads to a low-dimensional, unified representation of word images and strings, resulting in a method that allows one to perform either image and text searches, as well as image transcription, in a unified framework. We evaluate our methods on different public datasets of both handwritten documents and natural images showing results comparable or better than the state-of-the-art on spotting and recognition tasks.
Address
Corporate Author Thesis Ph.D. thesis
Publisher Ediciones Graficas Rey Place of Publication Editor Ernest Valveny;Alicia Fornes
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference
Notes DAG; 600.077 Approved no
Call Number Admin @ si @ Alm2014 Serial 2572
Permanent link to this record
 
 
Author Chris Bahnsen; David Vazquez; Antonio Lopez; Thomas B. Moeslund
Title (down) Learning to Remove Rain in Traffic Surveillance by Using Synthetic Data Type Conference Article
Year 2019 Publication 14th International Conference on Computer Vision Theory and Applications Abbreviated Journal
Volume Issue Pages 123-130
Keywords Rain Removal; Traffic Surveillance; Image Denoising
Abstract Rainfall is a problem in automated traffic surveillance. Rain streaks occlude the road users and degrade the overall visibility which in turn decrease object detection performance. One way of alleviating this is by artificially removing the rain from the images. This requires knowledge of corresponding rainy and rain-free images. Such images are often produced by overlaying synthetic rain on top of rain-free images. However, this method fails to incorporate the fact that rain fall in the entire three-dimensional volume of the scene. To overcome this, we introduce training data from the SYNTHIA virtual world that models rain streaks in the entirety of a scene. We train a conditional Generative Adversarial Network for rain removal and apply it on traffic surveillance images from SYNTHIA and the AAU RainSnow datasets. To measure the applicability of the rain-removed images in a traffic surveillance context, we run the YOLOv2 object detection algorithm on the original and rain-removed frames. The results on SYNTHIA show an 8% increase in detection accuracy compared to the original rain image. Interestingly, we find that high PSNR or SSIM scores do not imply good object detection performance.
Address Praga; Czech Republic; February 2019
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference VISIGRAPP
Notes ADAS; 600.118 Approved no
Call Number Admin @ si @ BVL2019 Serial 3256
Permanent link to this record
 
 
Author Albert Clapes
Title (down) Learning to recognize human actions: from hand-crafted to deep-learning based visual representations Type Book Whole
Year 2019 Publication PhD Thesis, Universitat de Barcelona-CVC Abbreviated Journal
Volume Issue Pages
Keywords
Abstract Action recognition is a very challenging and important problem in computer vi­sion. Researchers working on this field aspire to provide computers with the abil­ ity to visually perceive human actions – that is, to observe, interpret, and under­ stand human-related events that occur in the physical environment merely from visual data. The applications of this technology are numerous: human-machine interaction, e-health, monitoring/surveillance, and content-based video retrieval, among others. Hand-crafted methods dominated the field until the apparition of the first successful deep learning-based action recognition works. Although ear­ lier deep-based methods underperformed with respect to hand-crafted approaches, these slowly but steadily improved to become state-of-the-art, eventually achieving better results than hand-crafted ones. Still, hand-crafted approaches can be advan­ tageous in certain scenarios, specially when not enough data is available to train very large deep models or simply to be combined with deep-based methods to fur­ ther boost the performance. Hence, showing how hand-crafted features can provide extra knowledge the deep networks are notable to easily learn about human actions.
This Thesis concurs in time with this change of paradigm and, hence, reflects it into two distinguished parts. In the first part, we focus on improving current suc­ cessful hand-crafted approaches for action recognition and we do so from three dif­ ferent perspectives. Using the dense trajectories framework as a backbone: first, we explore the use of multi-modal and multi-view input
data to enrich the trajectory de­ scriptors. Second, we focus on the classification part of action recognition pipelines and propose an ensemble learning approach, where each classifier leams from a dif­ferent set of local spatiotemporal features to then combine their outputs following an strategy based on the Dempster-Shaffer Theory. And third, we propose a novel hand-crafted feature extraction method that constructs a rnid-level feature descrip­ tion to better modellong-term spatiotemporal dynarnics within action videos. Moving to the second part of the Thesis, we start with a comprehensive study of the current deep-learning based action recognition methods. We review both fun­ damental and cutting edge methodologies reported during the last few years and introduce a taxonomy of deep-leaming methods dedicated to action recognition. In particular, we analyze and discuss how these handle
the temporal dimension of data. Last but not least, we propose a residual recurrent network for action recogni­ tion that naturally integrates all our previous findings in a powerful and prornising framework.
Address January 2019
Corporate Author Thesis Ph.D. thesis
Publisher Ediciones Graficas Rey Place of Publication Editor Sergio Escalera
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN 978-84-948531-2-8 Medium
Area Expedition Conference
Notes HUPBA Approved no
Call Number Admin @ si @ Cla2019 Serial 3219
Permanent link to this record
 
 
Author Swathikiran Sudhakaran; Sergio Escalera;Oswald Lanz
Title (down) Learning to Recognize Actions on Objects in Egocentric Video with Attention Dictionaries Type Journal Article
Year 2021 Publication IEEE Transactions on Pattern Analysis and Machine Intelligence Abbreviated Journal TPAMI
Volume Issue Pages
Keywords
Abstract We present EgoACO, a deep neural architecture for video action recognition that learns to pool action-context-object descriptors from frame level features by leveraging the verb-noun structure of action labels in egocentric video datasets. The core component of EgoACO is class activation pooling (CAP), a differentiable pooling operation that combines ideas from bilinear pooling for fine-grained recognition and from feature learning for discriminative localization. CAP uses self-attention with a dictionary of learnable weights to pool from the most relevant feature regions. Through CAP, EgoACO learns to decode object and scene context descriptors from video frame features. For temporal modeling in EgoACO, we design a recurrent version of class activation pooling termed Long Short-Term Attention (LSTA). LSTA extends convolutional gated LSTM with built-in spatial attention and a re-designed output gate. Action, object and context descriptors are fused by a multi-head prediction that accounts for the inter-dependencies between noun-verb-action structured labels in egocentric video datasets. EgoACO features built-in visual explanations, helping learning and interpretation. Results on the two largest egocentric action recognition datasets currently available, EPIC-KITCHENS and EGTEA, show that by explicitly decoding action-context-object descriptors, EgoACO achieves state-of-the-art recognition performance.
Address
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference
Notes HUPBA; no proj Approved no
Call Number Admin @ si @ SEL2021 Serial 3656
Permanent link to this record
 
 
Author Pau Riba; Adria Molina; Lluis Gomez; Oriol Ramos Terrades; Josep Llados
Title (down) Learning to Rank Words: Optimizing Ranking Metrics for Word Spotting Type Conference Article
Year 2021 Publication 16th International Conference on Document Analysis and Recognition Abbreviated Journal
Volume 12822 Issue Pages 381–395
Keywords
Abstract In this paper, we explore and evaluate the use of ranking-based objective functions for learning simultaneously a word string and a word image encoder. We consider retrieval frameworks in which the user expects a retrieval list ranked according to a defined relevance score. In the context of a word spotting problem, the relevance score has been set according to the string edit distance from the query string. We experimentally demonstrate the competitive performance of the proposed model on query-by-string word spotting for both, handwritten and real scene word images. We also provide the results for query-by-example word spotting, although it is not the main focus of this work.
Address Lausanne; Suissa; September 2021
Corporate Author Thesis
Publisher Place of Publication Editor
Language Summary Language Original Title
Series Editor Series Title Abbreviated Series Title
Series Volume Series Issue Edition
ISSN ISBN Medium
Area Expedition Conference ICDAR
Notes DAG; 600.121; 600.140; 110.312 Approved no
Call Number Admin @ si @ RMG2021 Serial 3572
Permanent link to this record