Data augmentation is an important component in the robustness evaluation of models in natural language processing (NLP) and in enhancing the diversity of the data they are trained on. In this paper, we present NL-Augmenter, a new participatory Python-based natural language augmentation framework which supports the creation of both transformations (modifications to the data) and filters (data splits according to specific features). We describe the framework and an initial set of 117 transformations and 23 filters for a variety of natural language tasks. We demonstrate the efficacy of NL-Augmenter by using several of its transformations to analyze the robustness of popular natural language models. The infrastructure, datacards and robustness analysis results are available publicly on the NL-Augmenter repository (this https URL).

Superpixel segmentation consists of partitioning images into regions composed of similar and connected pixels. Its methods have been widely used in many computer vision applications since it allows for reducing the workload, removing redundant information, and preserving regions with meaningful features. Due to the rapid progress in this area, the literature fails to catch up on more recent works among the compared ones and to categorize the methods according to all existing strategies. This work fills this gap by presenting a comprehensive review with new taxonomy for superpixel segmentation, in which methods are classified according to their processing steps and processing levels of image features. We revisit the recent and popular literature according to our taxonomy and evaluate 20 strategies based on nine criteria: connectivity, compactness, delineation, control over the number of superpixels, color homogeneity, robustness, running time, stability, and visual quality. Our experiments show the trends of each approach in pixel clustering and discuss individual trade-offs. Finally, we provide a new benchmark for superpixel assessment, available at this https URL

With the growing amount of inappropriate content on the Internet, such as pornography, arises the need to detect and filter such material. The reason for this is given by the fact that such content is often prohibited in certain environments (e.g., schools and workplaces) or for certain publics (e.g., children). In recent years, many works have been mainly focused on detecting pornographic images and videos based on visual content, particularly on the detection of skin color. Although these approaches provide good results, they generally have the disadvantage of a high false positive rate since not all images with large areas of skin exposure are necessarily pornographic images, such as people wearing swimsuits or images related to sports. Local feature based approaches with Bag-of-Words models (BoW) have been successfully applied to visual recognition tasks in the context of pornography detection. Even though existing methods provide promising results, they use local feature descriptors that require a high computational processing time yielding high-dimensional vectors. In this work, we propose an approach for pornography detection based on local binary feature extraction and BossaNova image representation, a BoW model extension that preserves more richly the visual information. Moreover, we propose two approaches for video description based on the combination of mid-level representations namely BossaNova Video Descriptor (BNVD) and BoW Video Descriptor (BoW-VD). The proposed techniques are promising, achieving an accuracy of 92.40%, thus reducing the classification error by 16% over the current state-of-the-art local features approach on the Pornography dataset.

Research in the Vision and Language area encompasses challenging topics that seek to connect visual and textual information. When the visual information is related to videos, this takes us into Video-Text Research, which includes several challenging tasks such as video question answering, video summarization with natural language, and video-to-text and text-to-video conversion. This paper reviews the video-to-text problem, in which the goal is to associate an input video with its textual description. This association can be mainly made by retrieving the most relevant descriptions from a corpus or generating a new one given a context video. These two ways represent essential tasks for Computer Vision and Natural Language Processing communities, called text retrieval from video task and video captioning/description task. These two tasks are substantially more complex than predicting or retrieving a single sentence from an image. The spatiotemporal information present in videos introduces diversity and complexity regarding the visual content and the structure of associated language descriptions. This review categorizes and describes the state-of-the-art techniques for the video-to-text problem. It covers the main video-to-text methods and the ways to evaluate their performance. We analyze twenty-six benchmark datasets, showing their drawbacks and strengths for the problem requirements. We also show the progress that researchers have made on each dataset, we cover the challenges in the field, and we discuss future research directions.

Citizen science projects engage people in activities that are part of a scientific research effort. On multi-project citizen science platforms, scientists can create projects consisting of tasks. Volunteers, in turn, participate in executing the project's tasks. Such type of platforms seeks to connect volunteers and scientists' projects, adding value to both. However, little is known about volunteer's cross-project engagement patterns and the benefits of such patterns for scientists and volunteers. This work proposes a Goal, Question, and Metric (GQM) approach to analyse volunteers' cross-project task execution patterns and employs the Semiotic Inspection Method (SIM) to analyse the communicability of the platform's cross-project features. In doing so, it investigates what are the features of platforms to foster volunteers' cross-project engagement, to what extent multi-project platforms facilitate the attraction of volunteers to perform tasks in new projects, and to what extent multi-project participation increases engagement on the platforms. Results from analyses on real platforms show that volunteers tend to explore multiple projects, but they perform tasks regularly in just a few of them; few projects attract much attention from volunteers; volunteers recruited from other projects on the platform tend to get more engaged than those recruited outside the platform. System inspection shows that platforms still lack personalised and explainable recommendations of projects and tasks. The findings are translated into useful claims about how to design and manage multi-project platforms.

In recent years, deep learning has achieved formidable results in creative computing. When it comes to music, one viable model for music generation are Transformer based models. However, while transformers models are popular for music generation, they often rely on annotated structural information. In this work, we inquire if the off-the-shelf Music Transformer models perform just as well on structural similarity metrics using only unannotated MIDI information. We show that a slight tweak to the most common representation yields small but significant improvements. We also advocate that searching for better unannotated musical representations is more cost-effective than producing large amounts of curated and annotated data.

Software bots connect users and tools, streamlining the pull request review process in social coding platforms. However, bots can introduce information overload into developers' communication. Information overload is especially problematic for newcomers, who are still exploring the project and may feel overwhelmed by the number of messages. Inspired by the literature of other domains, we designed and evaluated FunnelBot, a bot that acts as a mediator between developers and other bots in the repository. We conducted a within-subject study with 25 newcomers to capture their perceptions and preferences. Our results provide insights for bot developers who want to mitigate noise and create bots for supporting newcomers, laying a foundation for designing better bots.

This work presents an approach for using GitHub classroom as a shared, structured, and persistent repository to support project-based courses at the Software Engineering Undergraduate program at PUC Minas, in Brazil. We discuss the needs of the different stakeholders that guided the development of the approach. Results on the perceptions of professors and students show that the approach brings benefits. Besides the lessons learned, we present insights on improving the education of the next generation of software engineers by employing metrics to monitor skill development, verifying student work portfolios, and employing tooling support in project-based courses.

State-of-the-art (SOTA) object detection methods have succeeded in several applications at the price of relying on heavyweight neural networks, which makes them inefficient and inviable for many applications with computational resource constraints. This work presents a method to build a Convolutional Neural Network (CNN) layer by layer for object detection from user-drawn markers on discriminative regions of representative images. We address the detection of Schistosomiasis mansoni eggs in microscopy images of fecal samples, and the detection of ships in satellite images as application examples. We could create a flyweight CNN without backpropagation from very few input images. Our method explores a recent methodology, Feature Learning from Image Markers (FLIM), to build convolutional feature extractors (encoders) from marker pixels. We extend FLIM to include a single-layer adaptive decoder, whose weights vary with the input image -- a concept never explored in CNNs. Our CNN weighs thousands of times less than SOTA object detectors, being suitable for CPU execution and showing superior or equivalent performance to three methods in five measures.

Salient Object Detection (SOD) with deep learning often requires substantial computational resources and large annotated datasets, making it impractical for resource-constrained applications. Lightweight models address computational demands but typically strive in complex and scarce labeled-data scenarios. Feature Learning from Image Markers (FLIM) learns an encoder's convolutional kernels among image patches extracted from discriminative regions marked on a few representative images, dismissing large annotated datasets, pretraining, and backpropagation. Such a methodology exploits information redundancy commonly found in biomedical image applications. This study presents methods to learn dilated-separable convolutional kernels and multi-dilation layers without backpropagation for FLIM networks. It also proposes a novel network simplification method to reduce kernel redundancy and encoder size. By combining a FLIM encoder with an adaptive decoder, a concept recently introduced to estimate a pointwise convolution per image, this study presents very efficient (named flyweight) SOD models for biomedical images. Experimental results in challenging datasets demonstrate superior efficiency and effectiveness to lightweight models. By requiring significantly fewer parameters and floating-point operations, the results show competitive effectiveness to heavyweight models. These advances highlight the potential of FLIM networks for data-limited and resource-constrained applications with information redundancy.