Helio Pedrini

Valter Estevam, Rayson Laroca, Helio Pedrini et al.

The success of Zero-shot Action Recognition (ZSAR) methods is intrinsically related to the nature of semantic side information used to transfer knowledge, although this aspect has not been primarily investigated in the literature. This work introduces a new ZSAR method based on the relationships of actions-objects and actions-descriptive sentences. We demonstrate that representing all object classes using descriptive sentences generates an accurate object-action affinity estimation when a paraphrase estimation method is used as an embedder. We also show how to estimate probabilities over the set of action classes based only on a set of sentences without hard human labeling. In our method, the probabilities from these two global classifiers (i.e., which use features computed over the entire video) are combined, producing an efficient transfer knowledge model for action classification. Our results are state-of-the-art in the Kinetics-400 dataset and are competitive on UCF-101 under the ZSAR evaluation. Our code is available at https://github.com/valterlej/objsentzsar

Valter Estevam, Rayson Laroca, Helio Pedrini et al.

This paper proposes a novel Zero-Shot Action Recognition~(ZSAR) method based on contrastive learning. In ZSAR, we aim to classify examples from classes that were missing during training. Two well-known problems remain in ZSAR: the semantic gap and the domain shift. A semantic gap occurs because label representations come from the textual domain (i.e., language models) and must be associated with visual representations (i.e., CNNs, RNNs, transformer-based). This multimodal nature implies that the semantic properties of the two spaces are not identical. On the other hand, the domain shift arises from differences between the training and test sets and is inherent to ZSAR once the test set is unknown. One of the most promising methods to address both issues is learning joint embedding spaces. Therefore, we propose a new model that encodes videos and sentences in a joint embedding space, trained by aligning videos with their natural-language descriptions. We design an automatic negative sampling procedure to augment the training dataset and generate unpaired data, i.e., visual appearance and unrelated descriptions. Our results are state-of-the-art on the UCF-101 and Kinetics-400 datasets under several split configurations. Our code is available at https://github.com/valterlej/cezsar.

Alexandre Lopes, Fernando Pereira dos Santos, Diulhio de Oliveira et al.

Deep neural networks have consistently represented the state of the art in most computer vision problems. In these scenarios, larger and more complex models have demonstrated superior performance to smaller architectures, especially when trained with plenty of representative data. With the recent adoption of Vision Transformer (ViT) based architectures and advanced Convolutional Neural Networks (CNNs), the total number of parameters of leading backbone architectures increased from 62M parameters in 2012 with AlexNet to 7B parameters in 2024 with AIM-7B. Consequently, deploying such deep architectures faces challenges in environments with processing and runtime constraints, particularly in embedded systems. This paper covers the main model compression techniques applied for computer vision tasks, enabling modern models to be used in embedded systems. We present the characteristics of compression subareas, compare different approaches, and discuss how to choose the best technique and expected variations when analyzing it on various embedded devices. We also share codes to assist researchers and new practitioners in overcoming initial implementation challenges for each subarea and present trends for Model Compression. Case studies for compression models are available at \href{https://github.com/venturusbr/cv-model-compression}{https://github.com/venturusbr/cv-model-compression}.

Gabriel Oliveira dos Santos, Diego A. B. Moreira, Alef Iury Ferreira et al.

This work introduces CAPIVARA, a cost-efficient framework designed to enhance the performance of multilingual CLIP models in low-resource languages. While CLIP has excelled in zero-shot vision-language tasks, the resource-intensive nature of model training remains challenging. Many datasets lack linguistic diversity, featuring solely English descriptions for images. CAPIVARA addresses this by augmenting text data using image captioning and machine translation to generate multiple synthetic captions in low-resource languages. We optimize the training pipeline with LiT, LoRA, and gradient checkpointing to alleviate the computational cost. Through extensive experiments, CAPIVARA emerges as state of the art in zero-shot tasks involving images and Portuguese texts. We show the potential for significant improvements in other low-resource languages, achieved by fine-tuning the pre-trained multilingual CLIP using CAPIVARA on a single GPU for 2 hours. Our model and code is available at https://github.com/hiaac-nlp/CAPIVARA.

Diego A. B. Moreira, Alef Iury Ferreira, Jhessica Silva et al.

Despite significant advancements and pervasive use of vision-language models, a paucity of studies has addressed their ethical implications. These models typically require extensive training data, often from hastily reviewed text and image datasets, leading to highly imbalanced datasets and ethical concerns. Additionally, models initially trained in English are frequently fine-tuned for other languages, such as the CLIP model, which can be expanded with more data to enhance capabilities but can add new biases. The CAPIVARA, a CLIP-based model adapted to Portuguese, has shown strong performance in zero-shot tasks. In this paper, we evaluate four different types of discriminatory practices within visual-language models and introduce FairPIVARA, a method to reduce them by removing the most affected dimensions of feature embeddings. The application of FairPIVARA has led to a significant reduction of up to 98% in observed biases while promoting a more balanced word distribution within the model. Our model and code are available at: https://github.com/hiaac-nlp/FairPIVARA.

George Corrêa de Araújo, Helio Pedrini

Single image super-resolution (SISR) is the process of obtaining one high-resolution version of a low-resolution image by increasing the number of pixels per unit area. This method has been actively investigated by the research community, due to the wide variety of real-world problems where it can be applied, from aerial and satellite imaging to compressed image and video enhancement. Despite the improvements achieved by deep learning in the field, the vast majority of the used networks are based on traditional convolutions, with the solutions focusing on going deeper and/or wider, and innovations coming from jointly employing successful concepts from other fields. In this work, we decided to step up from the traditional convolutions and adopt the concept of capsules. Since their overwhelming results both in image classification and segmentation problems, we question how suitable they are for SISR. We also verify that different solutions share most of their configurations, and argue that this trend leads to fewer explorations of network varieties. During our experiments, we check various strategies to improve results, ranging from new and different loss functions to changes in the capsule layers. Our network achieved good results with fewer convolutional-based layers, showing that capsules might be a concept worth applying in the image super-resolution problem.

Bruno Souza, Marius Aasan, Helio Pedrini et al.

The intersection of vision and language is of major interest due to the increased focus on seamless integration between recognition and reasoning. Scene graphs (SGs) have emerged as a useful tool for multimodal image analysis, showing impressive performance in tasks such as Visual Question Answering (VQA). In this work, we demonstrate that despite the effectiveness of scene graphs in VQA tasks, current methods that utilize idealized annotated scene graphs struggle to generalize when using predicted scene graphs extracted from images. To address this issue, we introduce the SelfGraphVQA framework. Our approach extracts a scene graph from an input image using a pre-trained scene graph generator and employs semantically-preserving augmentation with self-supervised techniques. This method improves the utilization of graph representations in VQA tasks by circumventing the need for costly and potentially biased annotated data. By creating alternative views of the extracted graphs through image augmentations, we can learn joint embeddings by optimizing the informational content in their representations using an un-normalized contrastive approach. As we work with SGs, we experiment with three distinct maximization strategies: node-wise, graph-wise, and permutation-equivariant regularization. We empirically showcase the effectiveness of the extracted scene graph for VQA and demonstrate that these approaches enhance overall performance by highlighting the significance of visual information. This offers a more practical solution for VQA tasks that rely on SGs for complex reasoning questions.

Thalles Santos Silva, Helio Pedrini, Adín Ramírez Rivera

We present Contextualized Local Visual Embeddings (CLoVE), a self-supervised convolutional-based method that learns representations suited for dense prediction tasks. CLoVE deviates from current methods and optimizes a single loss function that operates at the level of contextualized local embeddings learned from output feature maps of convolution neural network (CNN) encoders. To learn contextualized embeddings, CLoVE proposes a normalized mult-head self-attention layer that combines local features from different parts of an image based on similarity. We extensively benchmark CLoVE's pre-trained representations on multiple datasets. CLoVE reaches state-of-the-art performance for CNN-based architectures in 4 dense prediction downstream tasks, including object detection, instance segmentation, keypoint detection, and dense pose estimation.

Artur Jordao, George Correa de Araujo, Helena de Almeida Maia et al.

Pruning is a standard technique for reducing the computational cost of deep networks. Many advances in pruning leverage concepts from the Lottery Ticket Hypothesis (LTH). LTH reveals that inside a trained dense network exists sparse subnetworks (tickets) able to achieve similar accuracy (i.e., win the lottery - winning tickets). Pruning at initialization focuses on finding winning tickets without training a dense network. Studies on these concepts share the trend that subnetworks come from weight or filter pruning. In this work, we investigate LTH and pruning at initialization from the lens of layer pruning. First, we confirm the existence of winning tickets when the pruning process removes layers. Leveraged by this observation, we propose to discover these winning tickets at initialization, eliminating the requirement of heavy computational resources for training the initial (over-parameterized) dense network. Extensive experiments show that our winning tickets notably speed up the training phase and reduce up to 51% of carbon emission, an important step towards democratization and green Artificial Intelligence. Beyond computational benefits, our winning tickets exhibit robustness against adversarial and out-of-distribution examples. Finally, we show that our subnetworks easily win the lottery at initialization while tickets from filter removal (the standard structured LTH) hardly become winning tickets.

Igor Kenzo Ishikawa Oshiro Nakashima, Giovanna Vendramini, Helio Pedrini

Early and accurate diagnosis of COVID-19 is essential to control the rapid spread of the pandemic and mitigate sequelae in the population. Current diagnostic methods, such as RT-PCR, are effective but require time to provide results and can quickly overwhelm clinics, requiring individual laboratory analysis. Automatic detection methods have the potential to significantly reduce diagnostic time. To this end, learning-based methods using lung imaging have been explored. Although they require specialized hardware, automatic evaluation methods can be performed simultaneously, making diagnosis faster. Convolutional neural networks have been widely used to detect pneumonia caused by COVID-19 in lung images. This work describes an architecture based on 3D convolutional neural networks for detecting COVID-19 in computed tomography images. Despite the challenging scenario present in the dataset, the results obtained with our architecture demonstrated to be quite promising.

Thalles Silva, Helio Pedrini, Adín Ramírez Rivera

This paper introduces a novel approach to improving the training stability of self-supervised learning (SSL) methods by leveraging a non-parametric memory of seen concepts. The proposed method involves augmenting a neural network with a memory component to stochastically compare current image views with previously encountered concepts. Additionally, we introduce stochastic memory blocks to regularize training and enforce consistency between image views. We extensively benchmark our method on many vision tasks, such as linear probing, transfer learning, low-shot classification, and image retrieval on many datasets. The experimental results consolidate the effectiveness of the proposed approach in achieving stable SSL training without additional regularizers while learning highly transferable representations and requiring less computing time and resources.

Alexandre Lopes, Roberto Souza, Helio Pedrini

Depth Estimation plays a crucial role in recent applications in robotics, autonomous vehicles, and augmented reality. These scenarios commonly operate under constraints imposed by computational power. Stereo image pairs offer an effective solution for depth estimation since it only needs to estimate the disparity of pixels in image pairs to determine the depth in a known rectified system. Due to the difficulty in acquiring reliable ground-truth depth data across diverse scenarios, self-supervised techniques emerge as a solution, particularly when large unlabeled datasets are available. We propose a novel self-supervised convolutional approach that outperforms existing state-of-the-art Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) while balancing computational cost. The proposed CCNeXt architecture employs a modern CNN feature extractor with a novel windowed epipolar cross-attention module in the encoder, complemented by a comprehensive redesign of the depth estimation decoder. Our experiments demonstrate that CCNeXt achieves competitive metrics on the KITTI Eigen Split test data while being 10.18$\times$ faster than the current best model and achieves state-of-the-art results in all metrics in the KITTI Eigen Split Improved Ground Truth and Driving Stereo datasets when compared to recently proposed techniques. To ensure complete reproducibility, our project is accessible at \href{https://github.com/alelopes/CCNext}{\texttt{https://github.com/alelopes/CCNext}}.

Valter Estevam, Rayson Laroca, David Menotti et al.

This paper presents a novel approach to Zero-Shot Action Recognition. Recent works have explored the detection and classification of objects to obtain semantic information from videos with remarkable performance. Inspired by them, we propose using video captioning methods to extract semantic information about objects, scenes, humans, and their relationships. To the best of our knowledge, this is the first work to represent both videos and labels with descriptive sentences. More specifically, we represent videos using sentences generated via video captioning methods and classes using sentences extracted from documents acquired through search engines on the Internet. Using these representations, we build a shared semantic space employing BERT-based embedders pre-trained in the paraphrasing task on multiple text datasets. The projection of both visual and semantic information onto this space is straightforward, as they are sentences, enabling classification using the nearest neighbor rule. We demonstrate that representing videos and labels with sentences alleviates the domain adaptation problem. Additionally, we show that word vectors are unsuitable for building the semantic embedding space of our descriptions. Our method outperforms the state-of-the-art performance on the UCF101 dataset by 3.3 p.p. in accuracy under the TruZe protocol and achieves competitive results on both the UCF101 and HMDB51 datasets under the conventional protocol (0/50\% - training/testing split). Our code is available at https://github.com/valterlej/zsarcap.

Valter Estevam, Rayson Laroca, Helio Pedrini et al.

We introduce a method to learn unsupervised semantic visual information based on the premise that complex events can be decomposed into simpler events and that these simple events are shared across several complex events. We first employ a clustering method to group representations producing a visual codebook. Then, we learn a dense representation by encoding the co-occurrence probability matrix for the codebook entries. This representation leverages the performance of the dense video captioning task in a scenario with only visual features. For example, we replace the audio signal in the BMT method and produce temporal proposals with comparable performance. Furthermore, we concatenate the visual representation with our descriptor in a vanilla transformer method to achieve state-of-the-art performance in the captioning subtask compared to the methods that explore only visual features, as well as a competitive performance with multi-modal methods. Our code is available at https://github.com/valterlej/dvcusi.

Alexandre Lopes, Rodrigo Nogueira, Roberto Lotufo et al.

Despite the widespread adoption of deep learning for machine translation, it is still expensive to develop high-quality translation models. In this work, we investigate the use of pre-trained models, such as T5 for Portuguese-English and English-Portuguese translation tasks using low-cost hardware. We explore the use of Portuguese and English pre-trained language models and propose an adaptation of the English tokenizer to represent Portuguese characters, such as diaeresis, acute and grave accents. We compare our models to the Google Translate API and MarianMT on a subset of the ParaCrawl dataset, as well as to the winning submission to the WMT19 Biomedical Translation Shared Task. We also describe our submission to the WMT20 Biomedical Translation Shared Task. Our results show that our models have a competitive performance to state-of-the-art models while being trained on modest hardware (a single 8GB gaming GPU for nine days). Our data, models and code are available at https://github.com/unicamp-dl/Lite-T5-Translation.

Leandro Stival, Ricardo da Silva Torres, Helio Pedrini

Satellites continuously generate massive volumes of data, particularly for Earth observation, including satellite image time series (SITS). However, most deep learning models are designed to process either entire images or complete time series sequences to extract meaningful features for downstream tasks. In this study, we propose a novel multimodal approach that leverages pixel-wise two-dimensional (2D) representations to encode visual property variations from SITS more effectively. Specifically, we generate recurrence plots from pixel-based vegetation index time series (NDVI, EVI, and SAVI) as an alternative to using raw pixel values, creating more informative representations. Additionally, we introduce PIxel-wise Multimodal Contrastive (PIMC), a new multimodal self-supervision approach that produces effective encoders based on two-dimensional pixel time series representations and remote sensing imagery (RSI). To validate our approach, we assess its performance on three downstream tasks: pixel-level forecasting and classification using the PASTIS dataset, and land cover classification on the EuroSAT dataset. Moreover, we compare our results to state-of-the-art (SOTA) methods on all downstream tasks. Our experimental results show that the use of 2D representations significantly enhances feature extraction from SITS, while contrastive learning improves the quality of representations for both pixel time series and RSI. These findings suggest that our multimodal method outperforms existing models in various Earth observation tasks, establishing it as a robust self-supervision framework for processing both SITS and RSI. Code avaliable on

Thales Sales Almeida, Rodrigo Nogueira, Helio Pedrini

The performance of large language models (LLMs) is deeply influenced by the quality and composition of their training data. While much of the existing work has centered on English, there remains a gap in understanding how to construct effective training corpora for other languages. We explore scalable methods for building web-based corpora for LLMs. We apply them to build a new 120B token corpus in Portuguese that achieves competitive results to an industrial-grade corpus. Using a continual pretraining setup, we study how different data selection and preprocessing strategies affect LLM performance when transitioning a model originally trained in English to another language. Our findings demonstrate the value of language-specific filtering pipelines, including classifiers for education, science, technology, engineering, and mathematics (STEM), as well as toxic content. We show that adapting a model to the target language leads to performance improvements, reinforcing the importance of high-quality, language-specific data. While our case study focuses on Portuguese, our methods are applicable to other languages, offering insights for multilingual LLM development.

Gustavo Bonil, Simone Hashiguti, Jhessica Silva et al.

With the advance of Artificial Intelligence (AI), Large Language Models (LLMs) have gained prominence and been applied in diverse contexts. As they evolve into more sophisticated versions, it is essential to assess whether they reproduce biases, such as discrimination and racialization, while maintaining hegemonic discourses. Current bias detection approaches rely mostly on quantitative, automated methods, which often overlook the nuanced ways in which biases emerge in natural language. This study proposes a qualitative, discursive framework to complement such methods. Through manual analysis of LLM-generated short stories featuring Black and white women, we investigate gender and racial biases. We contend that qualitative methods such as the one proposed here are fundamental to help both developers and users identify the precise ways in which biases manifest in LLM outputs, thus enabling better conditions to mitigate them. Results show that Black women are portrayed as tied to ancestry and resistance, while white women appear in self-discovery processes. These patterns reflect how language models replicate crystalized discursive representations, reinforcing essentialization and a sense of social immobility. When prompted to correct biases, models offered superficial revisions that maintained problematic meanings, revealing limitations in fostering inclusive narratives. Our results demonstrate the ideological functioning of algorithms and have significant implications for the ethical use and development of AI. The study reinforces the need for critical, interdisciplinary approaches to AI design and deployment, addressing how LLM-generated discourses reflect and perpetuate inequalities.

Thalles Silva, Helio Pedrini, Adín Ramírez Rivera

We present Self-Organizing Visual Prototypes (SOP), a new training technique for unsupervised visual feature learning. Unlike existing prototypical self-supervised learning (SSL) methods that rely on a single prototype to encode all relevant features of a hidden cluster in the data, we propose the SOP strategy. In this strategy, a prototype is represented by many semantically similar representations, or support embeddings (SEs), each containing a complementary set of features that together better characterize their region in space and maximize training performance. We reaffirm the feasibility of non-parametric SSL by introducing novel non-parametric adaptations of two loss functions that implement the SOP strategy. Notably, we introduce the SOP Masked Image Modeling (SOP-MIM) task, where masked representations are reconstructed from the perspective of multiple non-parametric local SEs. We comprehensively evaluate the representations learned using the SOP strategy on a range of benchmarks, including retrieval, linear evaluation, fine-tuning, and object detection. Our pre-trained encoders achieve state-of-the-art performance on many retrieval benchmarks and demonstrate increasing performance gains with more complex encoders.

Jhessica Silva, Diego A. B. Moreira, Gabriel O. dos Santos et al.

In Artificial Intelligence (AI), language models have gained significant importance due to the widespread adoption of systems capable of simulating realistic conversations with humans through text generation. Because of their impact on society, developing and deploying these language models must be done responsibly, with attention to their negative impacts and possible harms. In this scenario, the number of AI Ethics Tools (AIETs) publications has recently increased. These AIETs are designed to help developers, companies, governments, and other stakeholders establish trust, transparency, and responsibility with their technologies by bringing accepted values to guide AI's design, development, and use stages. However, many AIETs lack good documentation, examples of use, and proof of their effectiveness in practice. This paper presents a methodology for evaluating AIETs in language models. Our approach involved an extensive literature survey on 213 AIETs, and after applying inclusion and exclusion criteria, we selected four AIETs: Model Cards, ALTAI, FactSheets, and Harms Modeling. For evaluation, we applied AIETs to language models developed for the Portuguese language, conducting 35 hours of interviews with their developers. The evaluation considered the developers' perspective on the AIETs' use and quality in helping to identify ethical considerations about their model. The results suggest that the applied AIETs serve as a guide for formulating general ethical considerations about language models. However, we note that they do not address unique aspects of these models, such as idiomatic expressions. Additionally, these AIETs did not help to identify potential negative impacts of models for the Portuguese language.

Christian Massao Konishi, Helio Pedrini

The preservation of the Amazon Rainforest is one of the global priorities in combating climate change, protecting biodiversity, and safeguarding indigenous cultures. The Satellite-based Monitoring Project of Deforestation in the Brazilian Legal Amazon (PRODES), a project of the National Institute for Space Research (INPE), stands out as a fundamental initiative in this effort, annually monitoring deforested areas not only in the Amazon but also in other Brazilian biomes. Recently, machine learning models have been developed using PRODES data to support this effort through the comparative analysis of multitemporal satellite images, treating deforestation detection as a change detection problem. However, existing approaches present significant limitations: models evaluated in the literature still show unsatisfactory effectiveness, many do not incorporate modern architectures, such as those based on self-attention mechanisms, and there is a lack of methodological standardization that allows direct comparisons between different studies. In this work, we address these gaps by evaluating various change detection models in a unified dataset, including fully convolutional models and networks incorporating self-attention mechanisms based on Transformers. We investigate the impact of different pre- and post-processing techniques, such as filtering deforested areas predicted by the models based on the size of connected components, texture replacement, and image enhancements; we demonstrate that such approaches can significantly improve individual model effectiveness. Additionally, we test different strategies for combining the evaluated models to achieve results superior to those obtained individually, reaching an F1-score of 80.41%, a value comparable to other recent works in the literature.

George Corrêa de Araújo, Helena de Almeida Maia, Helio Pedrini

In this paper, we present the Scrapbook framework, a novel methodology designed to generate extensive datasets for probing the learned concepts of artificial intelligence (AI) models. The framework focuses on fundamental concepts such as object recognition, absolute and relative positions, and attribute identification. By generating datasets with a large number of questions about individual concepts and a wide linguistic variation, the Scrapbook framework aims to validate the model's understanding of these basic elements before tackling more complex tasks. Our experimental findings reveal that, while contemporary models demonstrate proficiency in recognizing and enumerating objects, they encounter challenges in comprehending positional information and addressing inquiries with additional constraints. Specifically, the MobileVLM-V2 model showed significant answer disagreements and plausible wrong answers, while other models exhibited a bias toward affirmative answers and struggled with questions involving geometric shapes and positional information, indicating areas for improvement in understanding and consistency. The proposed framework offers a valuable instrument for generating diverse and comprehensive datasets, which can be utilized to systematically assess and enhance the performance of AI models.

Gustavo Bonil, João Gondim, Marina dos Santos et al.

This study investigates how large language models, in particular LLaMA 3.2-3B, construct narratives about Black and white women in short stories generated in Portuguese. From 2100 texts, we applied computational methods to group semantically similar stories, allowing a selection for qualitative analysis. Three main discursive representations emerge: social overcoming, ancestral mythification and subjective self-realization. The analysis uncovers how grammatically coherent, seemingly neutral texts materialize a crystallized, colonially structured framing of the female body, reinforcing historical inequalities. The study proposes an integrated approach, that combines machine learning techniques with qualitative, manual discourse analysis.

Gabriel Bianchin de Oliveira, Helio Pedrini, Zanoni Dias

Various approaches utilizing Transformer architectures have achieved state-of-the-art results in Natural Language Processing (NLP). Based on this success, numerous architectures have been proposed for other types of data, such as in biology, particularly for protein sequences. Notably among these are the ESM2 architectures, pre-trained on billions of proteins, which form the basis of various state-of-the-art approaches in the field. However, the ESM2 architectures have a limitation regarding input size, restricting it to 1,022 amino acids, which necessitates the use of preprocessing techniques to handle sequences longer than this limit. In this paper, we present the long and quantized versions of the ESM2 architectures, doubling the input size limit to 2,048 amino acids.

Lucas David, Helio Pedrini, Zanoni Dias

Weakly Supervised Semantic Segmentation (WSSS) techniques explore individual regularization strategies to refine Class Activation Maps (CAMs). In this work, we first analyze complementary WSSS techniques in the literature, their segmentation properties, and the conditions in which they are most effective. Based on these findings, we devise two new techniques: P-NOC and CCAM-H. In the first, we promote the conjoint training of two adversarial CAM generating networks: the generator, which progressively learns to erase regions containing class-specific features, and a discriminator, which is refined to gradually shift its attention to new class discriminant features. In the latter, we employ the high quality pseudo-segmentation priors produced by P-NOC to guide the learning to saliency information in a weakly supervised fashion. Finally, we employ both pseudo-segmentation priors and pseudo-saliency proposals in the random walk procedure, resulting in higher quality pseudo-semantic segmentation masks, and competitive results with the state of the art.

Alexandre Lopes, Roberto Souza, Helio Pedrini

RGB-D data is essential for solving many problems in computer vision. Hundreds of public RGB-D datasets containing various scenes, such as indoor, outdoor, aerial, driving, and medical, have been proposed. These datasets are useful for different applications and are fundamental for addressing classic computer vision tasks, such as monocular depth estimation. This paper reviewed and categorized image datasets that include depth information. We gathered 203 datasets that contain accessible data and grouped them into three categories: scene/objects, body, and medical. We also provided an overview of the different types of sensors, depth applications, and we examined trends and future directions of the usage and creation of datasets containing depth data, and how they can be applied to investigate the development of generalizable machine learning models in the monocular depth estimation field.

Edson Bollis, Helena Maia, Helio Pedrini et al.

Citrus juices and fruits are commodities with great economic potential in the international market, but productivity losses caused by mites and other pests are still far from being a good mark. Despite the integrated pest mechanical aspect, only a few works on automatic classification have handled images with orange mite characteristics, which means tiny and noisy regions of interest. On the computational side, attention-based models have gained prominence in deep learning research, and, along with weakly supervised learning algorithms, they have improved tasks performed with some label restrictions. In agronomic research of pests and diseases, these techniques can improve classification performance while pointing out the location of mites and insects without specific labels, reducing deep learning development costs related to generating bounding boxes. In this context, this work proposes an attention-based activation map approach developed to improve the classification of tiny regions called Two-Weighted Activation Mapping, which also produces locations using feature map scores learned from class labels. We apply our method in a two-stage network process called Attention-based Multiple Instance Learning Guided by Saliency Maps. We analyze the proposed approach in two challenging datasets, the Citrus Pest Benchmark, which was captured directly in the field using magnifying glasses, and the Insect Pest, a large pest image benchmark. In addition, we evaluate and compare our models with weakly supervised methods, such as Attention-based Deep MIL and WILDCAT. The results show that our classifier is superior to literature methods that use tiny regions in their classification tasks, surpassing them in all scenarios by at least 16 percentage points. Moreover, our approach infers bounding box locations for salient insects, even training without any location labels.

Artur Jordao, Helio Pedrini

Pruning is a well-known mechanism for reducing the computational cost of deep convolutional networks. However, studies have shown the potential of pruning as a form of regularization, which reduces overfitting and improves generalization. We demonstrate that this family of strategies provides additional benefits beyond computational performance and generalization. Our analyses reveal that pruning structures (filters and/or layers) from convolutional networks increase not only generalization but also robustness to adversarial images (natural images with content modified). Such achievements are possible since pruning reduces network capacity and provides regularization, which have been proven effective tools against adversarial images. In contrast to promising defense mechanisms that require training with adversarial images and careful regularization, we show that pruning obtains competitive results considering only natural images (e.g., the standard and low-cost training). We confirm these findings on several adversarial attacks and architectures; thus suggesting the potential of pruning as a novel defense mechanism against adversarial images.

Rodolfo Quispe, Cuiling Lan, Wenjun Zeng et al.

Vehicle Re-Identification (V-ReID) is a critical task that associates the same vehicle across images from different camera viewpoints. Many works explore attribute clues to enhance V-ReID; however, there is usually a lack of effective interaction between the attribute-related modules and final V-ReID objective. In this work, we propose a new method to efficiently explore discriminative information from vehicle attributes (for instance, color and type). We introduce AttributeNet (ANet) that jointly extracts identity-relevant features and attribute features. We enable the interaction by distilling the ReID-helpful attribute feature and adding it into the general ReID feature to increase the discrimination power. Moreover, we propose a constraint, named Amelioration Constraint (AC), which encourages the feature after adding attribute features onto the general ReID feature to be more discriminative than the original general ReID feature. We validate the effectiveness of our framework on three challenging datasets. Experimental results show that our method achieves the state-of-the-art performance.

Marcus de Assis Angeloni, Helio Pedrini

Recently, we have seen an increase in the global facial recognition market size. Despite significant advances in face recognition technology with the adoption of convolutional neural networks, there are still open challenges, such as when there is makeup in the face. To address this challenge, we propose and evaluate the adoption of facial parts to fuse with current holistic representations. We propose two strategies of facial parts: one with four regions (left periocular, right periocular, nose and mouth) and another with three facial thirds (upper, middle and lower). Experimental results obtained in four public makeup face datasets and in a challenging cross-dataset protocol show that the fusion of deep features extracted of facial parts with holistic representation increases the accuracy of face verification systems and decreases the error rates, even without any retraining of the CNN models. Our proposed pipeline achieved competitive results for the four datasets (EMFD, FAM, M501 and YMU).

Diogo C. Luvizon, Gustavo Sutter P. Carvalho, Andreza A. dos Santos et al.

In the last few years, several works have tackled the problem of novel view synthesis from stereo images or even from a single picture. However, previous methods are computationally expensive, specially for high-resolution images. In this paper, we address the problem of generating a multiplane image (MPI) from a single high-resolution picture. We present the adaptive-MPI representation, which allows rendering novel views with low computational requirements. To this end, we propose an adaptive slicing algorithm that produces an MPI with a variable number of image planes. We present a new lightweight CNN for depth estimation, which is learned by knowledge distillation from a larger network. Occluded regions in the adaptive-MPI are inpainted also by a lightweight CNN. We show that our method is capable of producing high-quality predictions with one order of magnitude less parameters compared to previous approaches. The robustness of our method is evidenced on challenging pictures from the Internet.

Rodolfo Quispe, Helio Pedrini

Person Re-Identification is a challenging task that aims to retrieve all instances of a query image across a system of non-overlapping cameras. Due to the various extreme changes of view, it is common that local regions that could be used to match people are suppressed, which leads to a scenario where approaches have to evaluate the similarity of images based on less informative regions. In this work, we introduce the Top-DB-Net, a method based on Top DropBlock that pushes the network to learn to focus on the scene foreground, with special emphasis on the most task-relevant regions and, at the same time, encodes low informative regions to provide high discriminability. The Top-DB-Net is composed of three streams: (i) a global stream encodes rich image information from a backbone, (ii) the Top DropBlock stream encourages the backbone to encode low informative regions with high discriminative features, and (iii) a regularization stream helps to deal with the noise created by the dropping process of the second stream, when testing the first two streams are used. Vast experiments on three challenging datasets show the capabilities of our approach against state-of-the-art methods. Qualitative results demonstrate that our method exhibits better activation maps focusing on reliable parts of the input images.

Allan Pinto, Manuel A. Córdova, Luis G. L. Decker et al.

Stereo vision is a growing topic in computer vision due to the innumerable opportunities and applications this technology offers for the development of modern solutions, such as virtual and augmented reality applications. To enhance the user's experience in three-dimensional virtual environments, the motion parallax estimation is a promising technique to achieve this objective. In this paper, we propose an algorithm for generating parallax motion effects from a single image, taking advantage of state-of-the-art instance segmentation and depth estimation approaches. This work also presents a comparison against such algorithms to investigate the trade-off between efficiency and quality of the parallax motion effects, taking into consideration a multi-task learning network capable of estimating instance segmentation and depth estimation at once. Experimental results and visual quality assessment indicate that the PyD-Net network (depth estimation) combined with Mask R-CNN or FBNet networks (instance segmentation) can produce parallax motion effects with good visual quality.

Edson Bollis, Helio Pedrini, Sandra Avila

Pests and diseases are relevant factors for production losses in agriculture and, therefore, promote a huge investment in the prevention and detection of its causative agents. In many countries, Integrated Pest Management is the most widely used process to prevent and mitigate the damages caused by pests and diseases in citrus crops. However, its results are credited by humans who visually inspect the orchards in order to identify the disease symptoms, insects and mite pests. In this context, we design a weakly supervised learning process guided by saliency maps to automatically select regions of interest in the images, significantly reducing the annotation task. In addition, we create a large citrus pest benchmark composed of positive samples (six classes of mite species) and negative samples. Experiments conducted on two large datasets demonstrate that our results are very promising for the problem of pest and disease classification in the agriculture field.

Rodolfo Quispe, Darwin Ttito, Adín Ramírez Rivera et al.

Crowd scene analysis has received a lot of attention recently due to the wide variety of applications, for instance, forensic science, urban planning, surveillance and security. In this context, a challenging task is known as crowd counting, whose main purpose is to estimate the number of people present in a single image. A Multi-Stream Convolutional Neural Network is developed and evaluated in this work, which receives an image as input and produces a density map that represents the spatial distribution of people in an end-to-end fashion. In order to address complex crowd counting issues, such as extremely unconstrained scale and perspective changes, the network architecture utilizes receptive fields with different size filters for each stream. In addition, we investigate the influence of the two most common fashions on the generation of ground truths and propose a hybrid method based on tiny face detection and scale interpolation. Experiments conducted on two challenging datasets, UCF-CC-50 and ShanghaiTech, demonstrate that using our ground truth generation methods achieves superior results.

Valter Estevam, Helio Pedrini, David Menotti

Zero-Shot Action Recognition has attracted attention in the last years and many approaches have been proposed for recognition of objects, events and actions in images and videos. There is a demand for methods that can classify instances from classes that are not present in the training of models, especially in the complex problem of automatic video understanding, since collecting, annotating and labeling videos are difficult and laborious tasks. We have identified that there are many methods available in the literature, however, it is difficult to categorize which techniques can be considered state of the art. Despite the existence of some surveys about zero-shot action recognition in still images and experimental protocol, there is no work focused on videos. Therefore, we present a survey of the methods that comprise techniques to perform visual feature extraction and semantic feature extraction as well to learn the mapping between these features considering specifically zero-shot action recognition in videos. We also provide a complete description of datasets, experiments and protocols, presenting open issues and directions for future work, essential for the development of the computer vision research field.

Rodolfo Quispe, Helio Pedrini

Given a video or an image of a person acquired from a camera, person re-identification is the process of retrieving all instances of the same person from videos or images taken from a different camera with non-overlapping view. This task has applications in various fields, such as surveillance, forensics, robotics, multimedia. In this paper, we present a novel framework, named Saliency-Semantic Parsing Re-Identification (SSP-ReID), for taking advantage of the capabilities of both clues: saliency and semantic parsing maps, to guide a backbone convolutional neural network (CNN) to learn complementary representations that improves the results over the original backbones. The insight of fusing multiple clues is based on specific scenarios in which one response is better than another, thus favoring the combination of them to increase performance. Due to its definition, our framework can be easily applied to a wide variety of networks and, in contrast to other competitive methods, our training process follows simple and standard protocols. We present extensive evaluation of our approach through five backbones and three benchmarks. Experimental results demonstrate the effectiveness of our person re-identification framework. In addition, we combine our framework with re-ranking techniques to achieve state-of-the-art results on three benchmarks.

David Menotti, Giovani Chiachia, Allan Pinto et al.

Biometrics systems have significantly improved person identification and authentication, playing an important role in personal, national, and global security. However, these systems might be deceived (or "spoofed") and, despite the recent advances in spoofing detection, current solutions often rely on domain knowledge, specific biometric reading systems, and attack types. We assume a very limited knowledge about biometric spoofing at the sensor to derive outstanding spoofing detection systems for iris, face, and fingerprint modalities based on two deep learning approaches. The first approach consists of learning suitable convolutional network architectures for each domain, while the second approach focuses on learning the weights of the network via back-propagation. We consider nine biometric spoofing benchmarks --- each one containing real and fake samples of a given biometric modality and attack type --- and learn deep representations for each benchmark by combining and contrasting the two learning approaches. This strategy not only provides better comprehension of how these approaches interplay, but also creates systems that exceed the best known results in eight out of the nine benchmarks. The results strongly indicate that spoofing detection systems based on convolutional networks can be robust to attacks already known and possibly adapted, with little effort, to image-based attacks that are yet to come.