With the COVID-19 global pandemic, computerassisted diagnoses of medical images have gained a lot of attention, and robust methods of Semantic Segmentation of Computed Tomography (CT) turned highly desirable. Semantic Segmentation of CT is one of many research fields of automatic detection of Covid-19 and was widely explored since the Covid-19 outbreak. In the robotic field, S emantic S egmentation of organs and CTs are widely used in robots developed for surgery tasks. As new methods and new datasets are proposed quickly, it becomes apparent the necessity of providing an extensive evaluation of those methods. To provide a standardized comparison of different architectures across multiple recently proposed datasets, we propose in this paper an extensive benchmark of multiple encoders and decoders with a total of 120 architectures evaluated in five d atasets, w ith e ach dataset being validated through a five-fold c ross-validation strategy, totaling 3.000 experiments. To the best of our knowledge, this is the largest evaluation in number of encoders, decoders, and datasets proposed in the field o f C ovid-19 C T segmentation. I. INTRODUCTION As of late 2019, the world faces the worst pandemic in years, with the new coronavirus disease, COVID-19, becoming a threat worldwide [1]. According to the global case count from the Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU) (updated April 20th, 2021), there are a total of 196,743,788 cases identified a ll around the globe, with a total of 4,201,812 global deaths [2]. Early diagnosis is one of the most effective ways to fight against the virus [3], with automatic detection of Covid-19 presence in CT being highly desirable, and recent results showing effectiveness in diagnosing and identifying Covid-19 patients [4]. Semantic Segmentation [5] of CT is one of many research fields of automatic detection of Covid-19 and was widely explored since the Covid-19 outbreak [4]. In the robotic field, S emantic S egmentation o f o rgans a nd C Ts are widely used in robots developed for surgery tasks [6], [7]. Aiming to perform the segmentation of Covid-19 CTs, many studies apply Deep Learning techniques and Deep Neural Networks, achieving impressive results in the task [4]. Deep Neural Networks are widely applied in segmentation problems due to their great generalization capacity, learning to represent different classes of objects [8], [9].However, with new approaches being proposed quickly, an urgency aggravated by the global pandemic, the need

No abstract

In this paper, we propose a broad comparison between Fully Convolutional Networks (FCNs) and Mask Regionbased Convolutional Neural Networks (Mask-RCNNs) applied in the Salient Object Detection (SOD) context. Studies in the SOD literature usually explore architectures based in FCNs to detect salient regions and objects in visual scenes. However, besides the promising results achieved, FCNs showed issues in some challenging scenarios. Fairly recently studies in the SOD literature proposed the use of a Mask-RCNN approach to overcome such issues. However, there is no extensive comparison between the two networks in the SOD literature endorsing the effectiveness of Mask-RCNNs over FCN when segmenting salient objects. Aiming to effectively show the superiority of Mask-RCNNs over FCNs in the SOD context, we compare two variations of Mask-RCNNs with two variations of FCNs in eight datasets widely used in the literature and in four metrics. Our findings show that in this context Mask-RCNNs achieved an improvement on the F-measure up to 47% over FCNs.

In this paper, we propose a novel data augmentation technique (ANDA) applied to the Salient Object Detection (SOD) context. Standard data augmentation techniques proposed in the literature, such as image cropping, rotation, flipping, and resizing, only generate variations of the existing examples, providing a limited generalization. Our method has the novelty of creating new images, by combining an object with a new background while retaining part of its salience in this new context; To do so, the ANDA technique relies on the linear combination between labeled salient objects and new backgrounds, generated by removing the original salient object in a process known as image inpainting. Our proposed technique allows for more precise control of the object's position and size while preserving background information. Aiming to evaluate our proposed method, we trained multiple deep neural networks and compared the effect that our technique has in each one. We also compared our method with other data augmentation techniques. Our findings show that depending on the network improvement can be up to 14.1% in the F-measure and decay of up to 2.6% in the Mean Absolute Error.

No abstract