|
Raul Gomez. (2020). Exploiting the Interplay between Visual and Textual Data for Scene Interpretation (Dimosthenis Karatzas, Lluis Gomez, & Jaume Gibert, Eds.). Ph.D. thesis, Ediciones Graficas Rey, .
Abstract: Machine learning experimentation under controlled scenarios and standard datasets is necessary to compare algorithms performance by evaluating all of them in the same setup. However, experimentation on how those algorithms perform on unconstrained data and applied tasks to solve real world problems is also a must to ascertain how that research can contribute to our society.
In this dissertation we experiment with the latest computer vision and natural language processing algorithms applying them to multimodal scene interpretation. Particularly, we research on how image and text understanding can be jointly exploited to address real world problems, focusing on learning from Social Media data.
We address several tasks that involve image and textual information, discuss their characteristics and offer our experimentation conclusions. First, we work on detection of scene text in images. Then, we work with Social Media posts, exploiting the captions associated to images as supervision to learn visual features, which we apply to multimodal semantic image retrieval. Subsequently, we work with geolocated Social Media images with associated tags, experimenting on how to use the tags as supervision, on location sensitive image retrieval and on exploiting location information for image tagging. Finally, we work on a specific classification problem of Social Media publications consisting on an image and a text: Multimodal hate speech classification.
|
|
|
Sounak Dey. (2020). Mapping between Images and Conceptual Spaces: Sketch-based Image Retrieval (Josep Llados, & Umapada Pal, Eds.). Ph.D. thesis, Ediciones Graficas Rey, .
Abstract: This thesis presents several contributions to the literature of sketch based image retrieval (SBIR). In SBIR the first challenge we face is how to map two different domains to common space for effective retrieval of images, while tackling the different levels of abstraction people use to express their notion of objects around while sketching. To this extent we first propose a cross-modal learning framework that maps both sketches and text into a joint embedding space invariant to depictive style, while preserving semantics. Then we have also investigated different query types possible to encompass people's dilema in sketching certain world objects. For this we propose an approach for multi-modal image retrieval in multi-labelled images. A multi-modal deep network architecture is formulated to jointly model sketches and text as input query modalities into a common embedding space, which is then further aligned with the image feature space. This permits encoding the object-based features and its alignment with the query irrespective of the availability of the co-occurrence of different objects in the training set.
Finally, we explore the problem of zero-shot sketch-based image retrieval (ZS-SBIR), where human sketches are used as queries to conduct retrieval of photos from unseen categories. We importantly advance prior arts by proposing a novel ZS-SBIR scenario that represents a firm step forward in its practical application. The new setting uniquely recognises two important yet often neglected challenges of practical ZS-SBIR, (i) the large domain gap between amateur sketch and photo, and (ii) the necessity for moving towards large-scale retrieval. We first contribute to the community a novel ZS-SBIR dataset, QuickDraw-Extended. We also in this dissertation pave the path to the future direction of research in this domain.
|
|
|
Marc Masana. (2020). Lifelong Learning of Neural Networks: Detecting Novelty and Adapting to New Domains without Forgetting (Joost Van de Weijer, & Andrew Bagdanov, Eds.). Ph.D. thesis, Ediciones Graficas Rey, .
Abstract: Computer vision has gone through considerable changes in the last decade as neural networks have come into common use. As available computational capabilities have grown, neural networks have achieved breakthroughs in many computer vision tasks, and have even surpassed human performance in others. With accuracy being so high, focus has shifted to other issues and challenges. One research direction that saw a notable increase in interest is on lifelong learning systems. Such systems should be capable of efficiently performing tasks, identifying and learning new ones, and should moreover be able to deploy smaller versions of themselves which are experts on specific tasks. In this thesis, we contribute to research on lifelong learning and address the compression and adaptation of networks to small target domains, the incremental learning of networks faced with a variety of tasks, and finally the detection of out-of-distribution samples at inference time.
We explore how knowledge can be transferred from large pretrained models to more task-specific networks capable of running on smaller devices by extracting the most relevant information. Using a pretrained model provides more robust representations and a more stable initialization when learning a smaller task, which leads to higher performance and is known as domain adaptation. However, those models are too large for certain applications that need to be deployed on devices with limited memory and computational capacity. In this thesis we show that, after performing domain adaptation, some learned activations barely contribute to the predictions of the model. Therefore, we propose to apply network compression based on low-rank matrix decomposition using the activation statistics. This results in a significant reduction of the model size and the computational cost.
Like human intelligence, machine intelligence aims to have the ability to learn and remember knowledge. However, when a trained neural network is presented with learning a new task, it ends up forgetting previous ones. This is known as catastrophic forgetting and its avoidance is studied in continual learning. The work presented in this thesis extensively surveys continual learning techniques and presents an approach to avoid catastrophic forgetting in sequential task learning scenarios. Our technique is based on using ternary masks in order to update a network to new tasks, reusing the knowledge of previous ones while not forgetting anything about them. In contrast to earlier work, our masks are applied to the activations of each layer instead of the weights. This considerably reduces the number of parameters to be added for each new task. Furthermore, the analysis on a wide range of work on incremental learning without access to the task-ID, provides insight on current state-of-the-art approaches that focus on avoiding catastrophic forgetting by using regularization, rehearsal of previous tasks from a small memory, or compensating the task-recency bias.
Neural networks trained with a cross-entropy loss force the outputs of the model to tend toward a one-hot encoded vector. This leads to models being too overly confident when presented with images or classes that were not present in the training distribution. The capacity of a system to be aware of the boundaries of the learned tasks and identify anomalies or classes which have not been learned yet is key to lifelong learning and autonomous systems. In this thesis, we present a metric learning approach to out-of-distribution detection that learns the task at hand on an embedding space.
|
|
|
Lei Kang. (2020). Robust Handwritten Text Recognition in Scarce Labeling Scenarios: Disentanglement, Adaptation and Generation (Alicia Fornes, Marçal Rusiñol, & Mauricio Villegas, Eds.). Ph.D. thesis, Ediciones Graficas Rey, .
Abstract: Handwritten documents are not only preserved in historical archives but also widely used in administrative documents such as cheques and claims. With the rise of the deep learning era, many state-of-the-art approaches have achieved good performance on specific datasets for Handwritten Text Recognition (HTR). However, it is still challenging to solve real use cases because of the varied handwriting styles across different writers and the limited labeled data. Thus, both explorin a more robust handwriting recognition architectures and proposing methods to diminish the gap between the source and target data in an unsupervised way are
demanded.
In this thesis, firstly, we explore novel architectures for HTR, from Sequence-to-Sequence (Seq2Seq) method with attention mechanism to non-recurrent Transformer-based method. Secondly, we focus on diminishing the performance gap between source and target data in an unsupervised way. Finally, we propose a group of generative methods for handwritten text images, which could be utilized to increase the training set to obtain a more robust recognizer. In addition, by simply modifying the generative method and joining it with a recognizer, we end up with an effective disentanglement method to distill textual content from handwriting styles so as to achieve a generalized recognition performance.
We outperform state-of-the-art HTR performances in the experimental results among different scientific and industrial datasets, which prove the effectiveness of the proposed methods. To the best of our knowledge, the non-recurrent recognizer and the disentanglement method are the first contributions in the handwriting recognition field. Furthermore, we have outlined the potential research lines, which would be interesting to explore in the future.
|
|
|
Manuel Carbonell. (2020). Neural Information Extraction from Semi-structured Documents A (Alicia Fornes, Mauricio Villegas, & Josep Llados, Eds.). Ph.D. thesis, Ediciones Graficas Rey, .
Abstract: Sectors as fintech, legaltech or insurance process an inflow of millions of forms, invoices, id documents, claims or similar every day. Together with these, historical archives provide gigantic amounts of digitized documents containing useful information that needs to be stored in machine encoded text with a meaningful structure. This procedure, known as information extraction (IE) comprises the steps of localizing and recognizing text, identifying named entities contained in it and optionally finding relationships among its elements. In this work we explore multi-task neural models at image and graph level to solve all steps in a unified way. While doing so we find benefits and limitations of these end-to-end approaches in comparison with sequential separate methods. More specifically, we first propose a method to produce textual as well as semantic labels with a unified model from handwritten text line images. We do so with the use of a convolutional recurrent neural model trained with connectionist temporal classification to predict the textual as well as semantic information encoded in the images. Secondly, motivated by the success of this approach we investigate the unification of the localization and recognition tasks of handwritten text in full pages with an end-to-end model, observing benefits in doing so. Having two models that tackle information extraction subsequent task pairs in an end-to-end to end manner, we lastly contribute with a method to put them all together in a single neural network to solve the whole information extraction pipeline in a unified way. Doing so we observe some benefits and some limitations in the approach, suggesting that in certain cases it is beneficial to train specialized models that excel at a single challenging task of the information extraction process, as it can be the recognition of named entities or the extraction of relationships between them. For this reason we lastly study the use of the recently arrived graph neural network architectures for the semantic tasks of the information extraction process, which are recognition of named entities and relation extraction, achieving promising results on the relation extraction part.
|
|
|
Gabriel Villalonga, & Antonio Lopez. (2020). Co-Training for On-Board Deep Object Detection. ACCESS - IEEE Access, , 194441–194456.
Abstract: Providing ground truth supervision to train visual models has been a bottleneck over the years, exacerbated by domain shifts which degenerate the performance of such models. This was the case when visual tasks relied on handcrafted features and shallow machine learning and, despite its unprecedented performance gains, the problem remains open within the deep learning paradigm due to its data-hungry nature. Best performing deep vision-based object detectors are trained in a supervised manner by relying on human-labeled bounding boxes which localize class instances (i.e. objects) within the training images. Thus, object detection is one of such tasks for which human labeling is a major bottleneck. In this article, we assess co-training as a semi-supervised learning method for self-labeling objects in unlabeled images, so reducing the human-labeling effort for developing deep object detectors. Our study pays special attention to a scenario involving domain shift; in particular, when we have automatically generated virtual-world images with object bounding boxes and we have real-world images which are unlabeled. Moreover, we are particularly interested in using co-training for deep object detection in the context of driver assistance systems and/or self-driving vehicles. Thus, using well-established datasets and protocols for object detection in these application contexts, we will show how co-training is a paradigm worth to pursue for alleviating object labeling, working both alone and together with task-agnostic domain adaptation.
|
|
|
Hannes Mueller, Andre Groger, Jonathan Hersh, Andrea Matranga, & Joan Serrat. (2020). Monitoring War Destruction from Space: A Machine Learning Approach.
Abstract: Existing data on building destruction in conflict zones rely on eyewitness reports or manual detection, which makes it generally scarce, incomplete and potentially biased. This lack of reliable data imposes severe limitations for media reporting, humanitarian relief efforts, human rights monitoring, reconstruction initiatives, and academic studies of violent conflict. This article introduces an automated method of measuring destruction in high-resolution satellite images using deep learning techniques combined with data augmentation to expand training samples. We apply this method to the Syrian civil war and reconstruct the evolution of damage in major cities across the country. The approach allows generating destruction data with unprecedented scope, resolution, and frequency – only limited by the available satellite imagery – which can alleviate data limitations decisively.
|
|
|
Yi Xiao, Felipe Codevilla, Akhil Gurram, Onay Urfalioglu, & Antonio Lopez. (2020). Multimodal end-to-end autonomous driving. TITS - IEEE Transactions on Intelligent Transportation Systems, , 1–11.
Abstract: A crucial component of an autonomous vehicle (AV) is the artificial intelligence (AI) is able to drive towards a desired destination. Today, there are different paradigms addressing the development of AI drivers. On the one hand, we find modular pipelines, which divide the driving task into sub-tasks such as perception and maneuver planning and control. On the other hand, we find end-to-end driving approaches that try to learn a direct mapping from input raw sensor data to vehicle control signals. The later are relatively less studied, but are gaining popularity since they are less demanding in terms of sensor data annotation. This paper focuses on end-to-end autonomous driving. So far, most proposals relying on this paradigm assume RGB images as input sensor data. However, AVs will not be equipped only with cameras, but also with active sensors providing accurate depth information (e.g., LiDARs). Accordingly, this paper analyses whether combining RGB and depth modalities, i.e. using RGBD data, produces better end-to-end AI drivers than relying on a single modality. We consider multimodality based on early, mid and late fusion schemes, both in multisensory and single-sensor (monocular depth estimation) settings. Using the CARLA simulator and conditional imitation learning (CIL), we show how, indeed, early fusion multimodality outperforms single-modality.
|
|
|
Andres Mafla, Ruben Tito, Sounak Dey, Lluis Gomez, Marçal Rusiñol, Ernest Valveny, et al. (2021). Real-time Lexicon-free Scene Text Retrieval. PR - Pattern Recognition, 110, 107656.
Abstract: In this work, we address the task of scene text retrieval: given a text query, the system returns all images containing the queried text. The proposed model uses a single shot CNN architecture that predicts bounding boxes and builds a compact representation of spotted words. In this way, this problem can be modeled as a nearest neighbor search of the textual representation of a query over the outputs of the CNN collected from the totality of an image database. Our experiments demonstrate that the proposed model outperforms previous state-of-the-art, while offering a significant increase in processing speed and unmatched expressiveness with samples never seen at training time. Several experiments to assess the generalization capability of the model are conducted in a multilingual dataset, as well as an application of real-time text spotting in videos.
|
|
|
Lluis Gomez, Anguelos Nicolaou, Marçal Rusiñol, & Dimosthenis Karatzas. (2020). 12 years of ICDAR Robust Reading Competitions: The evolution of reading systems for unconstrained text understanding. In K. Alahari, & C.V. Jawahar (Eds.), Visual Text Interpretation – Algorithms and Applications in Scene Understanding and Document Analysis. Series on Advances in Computer Vision and Pattern Recognition. Springer.
|
|
|
Lluis Gomez, Dena Bazazian, & Dimosthenis Karatzas. (2020). Historical review of scene text detection research. In K. Alahari, & C.V. Jawahar (Eds.), Visual Text Interpretation – Algorithms and Applications in Scene Understanding and Document Analysis. Series on Advances in Computer Vision and Pattern Recognition. Springer.
|
|
|
Jon Almazan, Lluis Gomez, Suman Ghosh, Ernest Valveny, & Dimosthenis Karatzas. (2020). WATTS: A common representation of word images and strings using embedded attributes for text recognition and retrieval. In K. A. Analysis”, & C.V. Jawahar (Eds.), Visual Text Interpretation – Algorithms and Applications in Scene Understanding and Document Analysis. Series on Advances in Computer Vision and Pattern Recognition. Springer.
|
|
|
Raul Gomez, Yahui Liu, Marco de Nadai, Dimosthenis Karatzas, Bruno Lepri, & Nicu Sebe. (2020). Retrieval Guided Unsupervised Multi-domain Image to Image Translation. In 28th ACM International Conference on Multimedia.
Abstract: Image to image translation aims to learn a mapping that transforms an image from one visual domain to another. Recent works assume that images descriptors can be disentangled into a domain-invariant content representation and a domain-specific style representation. Thus, translation models seek to preserve the content of source images while changing the style to a target visual domain. However, synthesizing new images is extremely challenging especially in multi-domain translations, as the network has to compose content and style to generate reliable and diverse images in multiple domains. In this paper we propose the use of an image retrieval system to assist the image-to-image translation task. First, we train an image-to-image translation model to map images to multiple domains. Then, we train an image retrieval model using real and generated images to find images similar to a query one in content but in a different domain. Finally, we exploit the image retrieval system to fine-tune the image-to-image translation model and generate higher quality images. Our experiments show the effectiveness of the proposed solution and highlight the contribution of the retrieval network, which can benefit from additional unlabeled data and help image-to-image translation models in the presence of scarce data.
|
|
|
Zhengying Liu, Zhen Xu, Shangeth Rajaa, Meysam Madadi, Julio C. S. Jacques Junior, Sergio Escalera, et al. (2020). Towards Automated Deep Learning: Analysis of the AutoDL challenge series 2019. In Proceedings of Machine Learning Research (Vol. 123, pp. 242–252).
Abstract: We present the design and results of recent competitions in Automated Deep Learning (AutoDL). In the AutoDL challenge series 2019, we organized 5 machine learning challenges: AutoCV, AutoCV2, AutoNLP, AutoSpeech and AutoDL. The first 4 challenges concern each a specific application domain, such as computer vision, natural language processing and speech recognition. At the time of March 2020, the last challenge AutoDL is still on-going and we only present its design. Some highlights of this work include: (1) a benchmark suite of baseline AutoML solutions, with emphasis on domains for which Deep Learning methods have had prior success (image, video, text, speech, etc); (2) a novel any-time learning framework, which opens doors for further theoretical consideration; (3) a repository of around 100 datasets (from all above domains) over half of which are released as public datasets to enable research on meta-learning; (4) analyses revealing that winning solutions generalize to new unseen datasets, validating progress towards universal AutoML solution; (5) open-sourcing of the challenge platform, the starting kit, the dataset formatting toolkit, and all winning solutions (All information available at {autodl.chalearn.org}).
|
|
|
Aymen Azaza, Joost Van de Weijer, Ali Douik, Javad Zolfaghari Bengar, & Marc Masana. (2020). Saliency from High-Level Semantic Image Features. SN - SN Computer Science, 1–12.
Abstract: Top-down semantic information is known to play an important role in assigning saliency. Recently, large strides have been made in improving state-of-the-art semantic image understanding in the fields of object detection and semantic segmentation. Therefore, since these methods have now reached a high-level of maturity, evaluation of the impact of high-level image understanding on saliency estimation is now feasible. We propose several saliency features which are computed from object detection and semantic segmentation results. We combine these features with a standard baseline method for saliency detection to evaluate their importance. Experiments demonstrate that the proposed features derived from object detection and semantic segmentation improve saliency estimation significantly. Moreover, they show that our method obtains state-of-the-art results on (FT, ImgSal, and SOD datasets) and obtains competitive results on four other datasets (ECSSD, PASCAL-S, MSRA-B, and HKU-IS).
|
|