Abstract: For humans, the recognition of objects is an almost instantaneous, precise and
extremely adaptable process. Furthermore, we have the innate capability to learn
new object classes from only few examples. The human brain lowers the complexity
of the incoming data by filtering out part of the information and only processing
those things that capture our attention. This, mixed with our biological predisposition to respond to certain shapes or colors, allows us to recognize in a simple
glance the most important or salient regions from an image. This mechanism can
be observed by analyzing on which parts of images subjects place attention; where
they fix their eyes when an image is shown to them. The most accurate way to
record this behavior is to track eye movements while displaying images.
Computational saliency estimation aims to identify to what extent regions or
objects stand out with respect to their surroundings to human observers. Saliency
maps can be used in a wide range of applications including object detection, image
and video compression, and visual tracking. The majority of research in the field has
focused on automatically estimating saliency maps given an input image. Instead, in
this thesis, we set out to incorporate saliency maps in an object recognition pipeline:
we want to investigate whether saliency maps can improve object recognition
results.
In this thesis, we identify several problems related to visual saliency estimation.
First, to what extent the estimation of saliency can be exploited to improve the
training of an object recognition model when scarce training data is available. To
solve this problem, we design an image classification network that incorporates
saliency information as input. This network processes the saliency map through a
dedicated network branch and uses the resulting characteristics to modulate the
standard bottom-up visual characteristics of the original image input. We will refer to this technique as saliency-modulated image classification (SMIC). In extensive
experiments on standard benchmark datasets for fine-grained object recognition,
we show that our proposed architecture can significantly improve performance,
especially on dataset with scarce training data.
Next, we address the main drawback of the above pipeline: SMIC requires an
explicit saliency algorithm that must be trained on a saliency dataset. To solve this,
we implement a hallucination mechanism that allows us to incorporate the saliency
estimation branch in an end-to-end trained neural network architecture that only
needs the RGB image as an input. A side-effect of this architecture is the estimation
of saliency maps. In experiments, we show that this architecture can obtain similar
results on object recognition as SMIC but without the requirement of ground truth
saliency maps to train the system.
Finally, we evaluated the accuracy of the saliency maps that occur as a sideeffect of object recognition. For this purpose, we use a set of benchmark datasets
for saliency evaluation based on eye-tracking experiments. Surprisingly, the estimated saliency maps are very similar to the maps that are computed from human
eye-tracking experiments. Our results show that these saliency maps can obtain
competitive results on benchmark saliency maps. On one synthetic saliency dataset
this method even obtains the state-of-the-art without the need of ever having seen
an actual saliency image for training.

Abstract: This book contains the proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) which was organized and sponsored by the Institute for Systems and Technologies of Information, Control and Communication (INSTICC), endorsed by the International Association for Pattern Recognition (IAPR), and in cooperation with the ACM Special Interest Group on Graphics and Interactive Techniques (SIGGRAPH), the European Association for Computer Graphics (EUROGRAPHICS), the EUROGRAPHICS Portuguese Chapter, the VRVis Center for Virtual Reality and Visualization Forschungs-GmbH, the French Association for Computer Graphics (AFIG), and the Society for Imaging Science and Technology (IS&T). The proceedings here published demonstrate new and innovative solutions and highlight technical problems in each field that are challenging and worthy of being disseminated to the interested research audiences. VISIGRAPP 2021 was organized to promote a discussion forum about the conference’s research topics between researchers, developers, manufacturers and end-users, and to establish guidelines in the development of more advanced solutions. This year VISIGRAPP was, exceptionally, held as a web-based event, due to the COVID-19 pandemic, from 8 – 10 February. We received a high number of paper submissions for this edition of VISIGRAPP, 371 in total, with contributions from 52 countries. This attests to the success and global dimension of VISIGRAPP. To evaluate each submission, we used a hierarchical process of double-blind evaluation where each paper was reviewed by two to six experts from the International Program Committee (IPC). The IPC selected for oral presentation and for publication as full papers 12 papers from GRAPP, 8 from HUCAPP, 11 papers from IVAPP, and 56 papers from VISAPP, which led to a result for the full-paper acceptance ratio of 24% and a high-quality program. Apart from the above full papers, the conference program also features 118 short papers and 67 poster presentations. We hope that these conference proceedings, which are submitted for indexation by Thomson Reuters Conference Proceedings Citation Index, SCOPUS, DBLP, Semantic Scholar, Google Scholar, EI and Microsoft Academic, will help the Computer Vision, Imaging, Visualization, Computer Graphics and Human-Computer Interaction communities to find interesting research work. Moreover, we are proud to inform that the program also includes three plenary keynote lectures, given by internationally distinguished researchers, namely Federico Tombari (Google and Technical University of Munich, Germany), Dieter Schmalstieg (Graz University of Technology, Austria) and Nathalie Henry Riche (Microsoft Research, United States), thus contributing to increase the overall quality of the conference and to provide a deeper understanding of the conference’s interest fields. Furthermore, a short list of the presented papers will be selected to be extended into a forthcoming book of VISIGRAPP Selected Papers to be published by Springer during 2021 in the CCIS series. Moreover, a short list of presented papers will be selected for publication of extended and revised versions in a special issue of the Springer Nature Computer Science journal. All papers presented at this conference will be available at the SCITEPRESS Digital Library. Three awards are delivered at the closing session, to recognize the best conference paper, the best student paper and the best poster for each of the four conferences. There is also an award for best industrial paper to be delivered at the closing session for VISAPP. We would like to express our thanks, first of all, to the authors of the technical papers, whose work and dedication made it possible to put together a program that we believe to be very exciting and of high technical quality. Next, we would like to thank the Area Chairs, all the members of the program committee and auxiliary reviewers, who helped us with their expertise and time. We would also like to thank the invited speakers for their invaluable contribution and for sharing their vision in their talks. Finally, we gratefully acknowledge the professional support of the INSTICC team for all organizational processes, especially given the need to introduce online streaming, forum management, direct messaging facilitation and other web-based activities in order to make it possible for VISIGRAPP 2021 authors to present their work and share ideas with colleagues in spite of the logistic difficulties caused by the current pandemic situation. We wish you all an exciting conference. We hope to meet you again for the next edition of VISIGRAPP, details of which are available at http://www. visigrapp.org

Abstract: In this work, we develop a generic framework that is able to handle the document retrieval problem in various scenarios such as searching for full page matches or retrieving the counterparts for specific document areas, focusing on their structural similarity or letting their visual resemblance to play a dominant role. Based on the spatial indexing technique, we propose to search for matches of local key-region pairs carrying both structural and visual information from the collection while a scheme allowing to adjust the relative contribution of structural and visual similarity is presented.
Based on the fact that the structure of documents is tightly linked with the distance among their elements, we firstly introduce an efficient detector named Distance Transform based Maximally Stable Extremal Regions (DTMSER). We illustrate that this detector is able to efficiently extract the structure of a document image as a dendrogram (hierarchical tree) of multi-scale key-regions that roughly correspond to letters, words and paragraphs. We demonstrate that, without benefiting from the structure information, the key-regions extracted by the DTMSER algorithm achieve better results comparing with state-of-the-art methods while much less amount of key-regions are employed.
We subsequently propose a pair-wise Bag of Words (BoW) framework to efficiently embed the explicit structure extracted by the DTMSER algorithm. We represent each document as a list of key-region pairs that correspond to the edges in the dendrogram where inclusion relationship is encoded. By employing those structural key-region pairs as the pooling elements for generating the histogram of features, the proposed method is able to encode the explicit inclusion relations into a BoW representation. The experimental results illustrate that the pair-wise BoW, powered by the embedded structural information, achieves remarkable improvement over the conventional BoW and spatial pyramidal BoW methods.
To handle various retrieval scenarios in one framework, we propose to directly query a series of key-region pairs, carrying both structure and visual information, from the collection. We introduce the spatial indexing techniques to the document retrieval community to speed up the structural relationship computation for key-region pairs. We firstly test the proposed framework in a full page retrieval scenario where structurally similar matches are expected. In this case, the pair-wise querying method achieves notable improvement over the BoW and spatial pyramidal BoW frameworks. Furthermore, we illustrate that the proposed method is also able to handle focused retrieval situations where the queries are defined as a specific interesting partial areas of the images. We examine our method on two types of focused queries: structure-focused and exact queries. The experimental results show that, the proposed generic framework obtains nearly perfect precision on both types of focused queries while it is the first framework able to tackle structure-focused queries, setting a new state of the art in the field.
Besides, we introduce a line verification method to check the spatial consistency among the matched key-region pairs. We propose a computationally efficient version of line verification through a two step implementation. We first compute tentative localizations of the query and subsequently employ them to divide the matched key-region pairs into several groups, then line verification is performed within each group while more precise bounding boxes are computed. We demonstrate that, comparing with the standard approach (based on RANSAC), the line verification proposed generally achieves much higher recall with slight loss on precision on specific queries.

Abstract: Text is the most common way of information sharing from ages. With recent development of personal images databases and handwritten historic manuscripts the demand for algorithms to make these databases accessible for browsing and indexing are in rise. Enabling search or understanding large collection of manuscripts or image databases needs fast and robust methods. Researchers have found different ways to represent cropped words for understanding and matching, which works well when words are already segmented. However there is no trivial way to extend these for non-segmented documents. In this thesis we explore different methods for text retrieval and recognition from unsegmented document and scene images. Two different ways of representation exist in literature, one uses a fixed length representation learned from cropped words and another a sequence of features of variable length. Throughout this thesis, we have studied both these representation for their suitability in segmentation free understanding of text. In the first part we are focused on segmentation free word spotting using a fixed length representation. We extended the use of the successful PHOC (Pyramidal Histogram of Character) representation to segmentation free retrieval. In the second part of the thesis, we explore sequence based features and finally, we propose a unified solution where the same framework can generate both kind of representations.

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.

Abstract: The curse of data labeling is a costly bottleneck in supervised deep learning, where large amounts of labeled data are needed to train intelligent systems. In onboard perception for autonomous driving, this cost corresponds to the labeling of raw data from sensors such as cameras, LiDARs, RADARs, etc. Therefore, synthetic data with automatically generated ground truth (labels) has aroused as a reliable alternative for training onboard perception models.
However, synthetic data commonly suffers from synth-to-real domain shift, i.e., models trained on the synthetic domain do not show their achievable accuracy when performing in the real world. This shift needs to be addressed by techniques falling in the realm of domain adaptation (DA).
The semi-supervised learning (SSL) paradigm can be followed to address DA. In this case, a model is trained using source data with labels (here synthetic) and leverages minimal knowledge from target data (here the real world) to generate pseudo-labels. These pseudo-labels help the training process to reduce the gap between the source and the target domains. In general, we can assume accessing both, pseudo-labels and a few amounts of human-provided labels for the target-domain data. However, the most interesting and challenging setting consists in assuming that we do not have human-provided labels at all. This setting is known as unsupervised domain adaptation (UDA). This PhD focuses on applying SSL to the UDA setting, for onboard visual tasks related to autonomous driving. We start by addressing the synth-to-real UDA problem on onboard vision-based object detection (pedestrians and cars), a critical task for autonomous driving and driving assistance. In particular, we propose to apply an SSL technique known as co-training, which we adapt to work with deep models that process a multi-modal input. The multi-modality consists of the visual appearance of the images (RGB) and their monocular depth estimation. The synthetic data we use as the source domain contains both, object bounding boxes and depth information. This prior knowledge is the
starting point for the co-training technique, which iteratively labels unlabeled real-world data and uses such pseudolabels (here bounding boxes with an assigned object class) to progressively improve the labeling results. Along this
process, two models collaborate to automatically label the images, in a way that one model compensates for the errors of the other, so avoiding error drift. While this automatic labeling process is done offline, the resulting pseudolabels can be used to train object detection models that must perform in real-time onboard a vehicle. We show that multi-modal co-training improves the labeling results compared to single-modal co-training, remaining competitive compared to human labeling.
Given the success of co-training in the context of object detection, we have also adapted this technique to a more crucial and challenging visual task, namely, onboard semantic segmentation. In fact, providing labels for a single image
can take from 30 to 90 minutes for a human labeler, depending on the content of the image. Thus, developing automatic labeling techniques for this visual task is of great interest to the automotive industry. In particular, the new co-training framework addresses synth-to-real UDA by an initial stage of self-training. Intermediate models arising from this stage are used to start the co-training procedure, for which we have elaborated an accurate collaboration policy between the two models performing the automatic labeling. Moreover, our co-training seamlessly leverages datasets from different synthetic domains. In addition, the co-training procedure is agnostic to the loss function used to train the semantic segmentation models which perform the automatic labeling. We achieve state-of-the-art results on publicly available benchmark datasets, again, remaining competitive compared to human labeling.
Finally, on the ground of our previous experience, we have designed and implemented a new SSL technique for UDA in the context of visual semantic segmentation. In this case, we mimic the labeling methodology followed by human labelers. In particular, rather than labeling full images at a time, categories of semantic classes are defined and only those are labeled in a labeling pass. In fact, different human labelers can become specialists in labeling different categories. Afterward, these per-category-labeled layers are combined to provide fully labeled images. Our technique is inspired by this methodology since we perform synth-to-real UDA per category, using the self-training stage previously developed as part of our co-training framework. The pseudo-labels obtained for each category are finally
fused to obtain fully automatically labeled images. In this context, we have also contributed to the development of a new photo-realistic synthetic dataset based on path-tracing rendering. Our new SSL technique seamlessly leverages publicly available synthetic datasets as well as this new one to obtain state-of-the-art results on synth-to-real UDA for semantic segmentation. We show that the new dataset allows us to reach better labeling accuracy than previously existing datasets, at the same time that it complements well them when combined. Moreover, we also show that the new human-inspired SSL technique outperforms co-training.