João Paulo Gois
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
João Paulo Gois has a total of 70 co-authors. Here are the top co-authors by number of shared publications:
- Charles Morphy Dias Santos
Centro de Ciências Naturais e Humanas, Universidade Federal do ABC, Santo André, Brazil
11 shared publications - Harlen Costa Batagelo
Center for Mathematics, Computing and Cognition, Federal University of ABC, Santo André, Brazil
7 shared publications - Hiroki Takahashi
Graduate School of Informatics and Engineering, The University of Electro-Communications, Chofu, Japan
6 shared publications - Fernanda Miyuki Yamada
Graduate School of Informatics and Engineering, The University of Electro-Communications, Chofu, Tokyo, Japan
5 shared publications - André Luiz Brandão
UFABC
4 shared publications Antonio Castelo
Universidade de São Paulo, Brazil
3 shared publications
Latest Publications
- Open Access
- 12 October 2026
- Published by Springer Nature in Marine Biology
- Vol. 173 (11) , 211
HideAndSeg: an AI-driven approach to monitoring benthic octopuses in the wild
Analyzing the behavior and body patterns of octopuses in nature is challenging due to their camouflage capabilities, changes in skin texture and color, body deformations, and frequent occlusions. Underwater lighting and turbidity further complicate these challenges. Addressing the lack of large-scale annotated datasets, we introduce HideAndSeg, a minimally supervised, AI-based tool for segmenting octopus videos. HideAndSeg integrates SAM2 with a custom-trained YOLOv11 object detector. First, the user provides point coordinates to generate the initial segmentation masks with SAM2. These masks serve as training data for the YOLO model. Our approach automates the pipeline by providing a bounding-box prompt to SAM2, eliminating manual intervention. We introduce two unsupervised metrics – temporal consistency and new component count – to quantitatively evaluate segmentation quality and guide mask refinement in the absence of real-world information used to train, validate, and test AI models. HideAndSeg reduces segmentation noise compared to the manually prompted approach. Our method can re-identify and segment the octopus even after complete occlusion, a scenario in which the manually prompted model fails. By reducing manual analysis in real-world scenarios, this work provides a practical tool that enables more efficient behavioral studies of wild octopuses.
- 12 October 2026
- Published by Springer Nature in Marine Biology
- Vol. 173 (11) , 211
Open AccessHideAndSeg: an AI-driven approach to monitoring benthic octopuses in the wild
Analyzing the behavior and body patterns of octopuses in nature is challenging due to their camouflage capabilities, changes in skin texture and color, body deformations, and frequent occlusions. Underwater lighting and turbidity further complicate these challenges. Addressing the lack of large-scale annotated datasets, we introduce HideAndSeg, a minimally supervised, AI-based tool for segmenting octopus videos. HideAndSeg integrates SAM2 with a custom-trained YOLOv11 object detector. First, the user provides point coordinates to generate the initial segmentation masks with SAM2. These masks serve as training data for the YOLO model. Our approach automates the pipeline by providing a bounding-box prompt to SAM2, eliminating manual intervention. We introduce two unsupervised metrics – temporal consistency and new component count – to quantitatively evaluate segmentation quality and guide mask refinement in the absence of real-world information used to train, validate, and test AI models. HideAndSeg reduces segmentation noise compared to the manually prompted approach. Our method can re-identify and segment the octopus even after complete occlusion, a scenario in which the manually prompted model fails. By reducing manual analysis in real-world scenarios, this work provides a practical tool that enables more efficient behavioral studies of wild octopuses.
- Open Access
- 9 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Motion Retargeting for 3D Avatars with Upper Limb Differences
Most text-to-3D avatar generation methods assume normative body structures and cannot represent limb differences or amputations. DreamAble introduced 3D avatar generation with upper limb differences but did not address their animation. This paper extends DreamAble to animation by incorporating adapted skeletal structures into conventional rigging and motion retargeting pipelines. Motion sequences from actors with normative body structures are retargeted to 3D avatars with upper limb differences. Our findings advance inclusive 3D avatar animation, with potential applications in gaming, social virtual reality, and medical training environments.
- 9 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Open AccessMotion Retargeting for 3D Avatars with Upper Limb Differences
Most text-to-3D avatar generation methods assume normative body structures and cannot represent limb differences or amputations. DreamAble introduced 3D avatar generation with upper limb differences but did not address their animation. This paper extends DreamAble to animation by incorporating adapted skeletal structures into conventional rigging and motion retargeting pipelines. Motion sequences from actors with normative body structures are retargeted to 3D avatars with upper limb differences. Our findings advance inclusive 3D avatar animation, with potential applications in gaming, social virtual reality, and medical training environments.
- Open Access
- 11 June 2026
- Published by Center for Open Science in EdArXiv
Pro não é noob: Introduzindo conceitos de jogos digitais na mesa de trabalho docente
O avanço dos jogos digitais na sociedade contemporânea contrasta com as barreiras à sua integração na prática pedagógica da educação básica brasileira. Diante do descompasso entre a formação docente tradicional e a realidade discente, marcada pelo consumo massivo de jogos em dispositivos móveis pessoais, este trabalho apresenta uma discussão sobre o papel do professor "Pro". Frente às limitações orçamentárias e à descontinuidade técnica de protótipos educativos acadêmicos, defende-se a apropriação crítica de jogos comerciais de prateleira. Para tanto, instrumentaliza-se o educador com ferramentas analíticas e metodológicas de game design, especificamente o Framework MDA, a Tétrade Elementar de Schell e o Documento de Design de Jogo (GDD) adaptado ao contexto escolar. Abordam-se também estratégias de gestão de infraestrutura, questões de autoria digital e os impactos da Inteligência Artificial Generativa e de tecnologias imersivas sob uma perspectiva ética de cocriação. Conclui-se que a fundamentação metodológica e a capacidade analítica do docente constituem competências perenes e insubstituíveis para o letramento digital, independentemente das contínuas transformações técnicas da indústria de jogos.
- 11 June 2026
- Published by Center for Open Science in EdArXiv
Open AccessPro não é noob: Introduzindo conceitos de jogos digitais na mesa de trabalho docente
O avanço dos jogos digitais na sociedade contemporânea contrasta com as barreiras à sua integração na prática pedagógica da educação básica brasileira. Diante do descompasso entre a formação docente tradicional e a realidade discente, marcada pelo consumo massivo de jogos em dispositivos móveis pessoais, este trabalho apresenta uma discussão sobre o papel do professor "Pro". Frente às limitações orçamentárias e à descontinuidade técnica de protótipos educativos acadêmicos, defende-se a apropriação crítica de jogos comerciais de prateleira. Para tanto, instrumentaliza-se o educador com ferramentas analíticas e metodológicas de game design, especificamente o Framework MDA, a Tétrade Elementar de Schell e o Documento de Design de Jogo (GDD) adaptado ao contexto escolar. Abordam-se também estratégias de gestão de infraestrutura, questões de autoria digital e os impactos da Inteligência Artificial Generativa e de tecnologias imersivas sob uma perspectiva ética de cocriação. Conclui-se que a fundamentação metodológica e a capacidade analítica do docente constituem competências perenes e insubstituíveis para o letramento digital, independentemente das contínuas transformações técnicas da indústria de jogos.
- 1 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computing in Science & Engineering
- Vol. 28 (2) , 81-88
Toward Inclusive Avatar Design With Limb Differences Through Artificial Intelligence
- 1 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computing in Science & Engineering
- Vol. 28 (2) , 81-88
Toward Inclusive Avatar Design With Limb Differences Through Artificial Intelligence
- 24 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Graphics and Applications
- Vol. 46 (2) , 112-119
The Rise of AI-Generated Anime Avatars: Trends, Challenges, and Opportunities
The rise of 3-D anime-style avatars in gaming, virtual reality, and other digital media has driven significant interest in automated generation methods capable of capturing their distinctive visual characteristics. These include stylized proportions, expressive features, and nonphotorealistic rendering. This article reviews the advancements and challenges in using deep learning in 3-D anime-style avatar generation. We analyze the strengths and limitations of these methods in capturing the aesthetics of anime characters and supporting customization and animation. In addition, we identify and discuss open problems in the field, such as difficulties in resolution and detail preservation, and constraints regarding the animation of hair and loose clothing. This article aims to provide a comprehensive overview of the current state-of-the-art and identify promising research directions for advancing 3-D anime-style avatar generation.
- 24 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Graphics and Applications
- Vol. 46 (2) , 112-119
The Rise of AI-Generated Anime Avatars: Trends, Challenges, and Opportunities
The rise of 3-D anime-style avatars in gaming, virtual reality, and other digital media has driven significant interest in automated generation methods capable of capturing their distinctive visual characteristics. These include stylized proportions, expressive features, and nonphotorealistic rendering. This article reviews the advancements and challenges in using deep learning in 3-D anime-style avatar generation. We analyze the strengths and limitations of these methods in capturing the aesthetics of anime characters and supporting customization and animation. In addition, we identify and discuss open problems in the field, such as difficulties in resolution and detail preservation, and constraints regarding the animation of hair and loose clothing. This article aims to provide a comprehensive overview of the current state-of-the-art and identify promising research directions for advancing 3-D anime-style avatar generation.
- Open Access
HideAndSeg: an AI-based tool with automated prompting for octopus segmentation in natural habitats
Analyzing octopuses in their natural habitats is challenging due to their camouflage capability, rapid changes in skin texture and color, non-rigid body deformations, and frequent occlusions, all of which are compounded by variable underwater lighting and turbidity. Addressing the lack of large-scale annotated datasets, this paper introduces HideAndSeg, a novel, minimally supervised AI-based tool for segmenting videos of octopuses. It establishes a quantitative baseline for this task. HideAndSeg integrates SAM2 with a custom-trained YOLOv11 object detector. First, the user provides point coordinates to generate the initial segmentation masks with SAM2. These masks serve as training data for the YOLO model. After that, our approach fully automates the pipeline by providing a bounding box prompt to SAM2, eliminating the need for further manual intervention. We introduce two unsupervised metrics - temporal consistency $DICE_t$ and new component count $NC_t$ - to quantitatively evaluate segmentation quality and guide mask refinement in the absence of ground-truth data, i.e., real-world information that serves to train, validate, and test AI models. Results show that HideAndSeg achieves satisfactory performance, reducing segmentation noise compared to the manually prompted approach. Our method can re-identify and segment the octopus even after periods of complete occlusion in natural environments, a scenario in which the manually prompted model fails. By reducing the need for manual analysis in real-world scenarios, this work provides a practical tool that paves the way for more efficient behavioral studies of wild cephalopods.Open AccessHideAndSeg: an AI-based tool with automated prompting for octopus segmentation in natural habitats
Analyzing octopuses in their natural habitats is challenging due to their camouflage capability, rapid changes in skin texture and color, non-rigid body deformations, and frequent occlusions, all of which are compounded by variable underwater lighting and turbidity. Addressing the lack of large-scale annotated datasets, this paper introduces HideAndSeg, a novel, minimally supervised AI-based tool for segmenting videos of octopuses. It establishes a quantitative baseline for this task. HideAndSeg integrates SAM2 with a custom-trained YOLOv11 object detector. First, the user provides point coordinates to generate the initial segmentation masks with SAM2. These masks serve as training data for the YOLO model. After that, our approach fully automates the pipeline by providing a bounding box prompt to SAM2, eliminating the need for further manual intervention. We introduce two unsupervised metrics - temporal consistency $DICE_t$ and new component count $NC_t$ - to quantitatively evaluate segmentation quality and guide mask refinement in the absence of ground-truth data, i.e., real-world information that serves to train, validate, and test AI models. Results show that HideAndSeg achieves satisfactory performance, reducing segmentation noise compared to the manually prompted approach. Our method can re-identify and segment the octopus even after periods of complete occlusion in natural environments, a scenario in which the manually prompted model fails. By reducing the need for manual analysis in real-world scenarios, this work provides a practical tool that paves the way for more efficient behavioral studies of wild cephalopods. - Open Access
- 30 September 2025
- Published by Sociedade Brasileira de Computação
- p. 1137-1148
Diptera-VR: A Gamified Virtual Entomology Lab for Science Education
Introduction: Games and Virtual Reality (VR) have proven to be powerful tools for science education, enabling immersive experiences that enhance practical learning. However, gamified virtual laboratories remain underexplored in entomology education, thereby limiting engagement and accessibility to scientific knowledge. Objective: To present Diptera-VR, a gamified Virtual Reality experience designed to simulate the environment and activities of an entomology laboratory at the Universidade Federal do ABC (UFABC), Brazil. The experience promotes science education and outreach through an interactive and immersive approach. Methodology: Diptera-VR was developed using 3D modeling and VR interaction techniques to create an engaging educational environment. Players can manipulate laboratory equipment, examine insect specimens, and perform tasks such as inspecting wings under a microscope, analyzing fossil material, and observing pinned specimens. The development process included the creation of digital assets and interaction design to ensure an accessible educational experience. Results: Diptera-VR is expected to bridge the gap between theoretical learning and scientific practice, increasing student engagement and improving knowledge retention. Additionally, it may contribute to the popularization of science in a context marked by a global environmental crisis and growing scientific denialism.
- 30 September 2025
- Published by Sociedade Brasileira de Computação
- p. 1137-1148
Open AccessDiptera-VR: A Gamified Virtual Entomology Lab for Science Education
Introduction: Games and Virtual Reality (VR) have proven to be powerful tools for science education, enabling immersive experiences that enhance practical learning. However, gamified virtual laboratories remain underexplored in entomology education, thereby limiting engagement and accessibility to scientific knowledge. Objective: To present Diptera-VR, a gamified Virtual Reality experience designed to simulate the environment and activities of an entomology laboratory at the Universidade Federal do ABC (UFABC), Brazil. The experience promotes science education and outreach through an interactive and immersive approach. Methodology: Diptera-VR was developed using 3D modeling and VR interaction techniques to create an engaging educational environment. Players can manipulate laboratory equipment, examine insect specimens, and perform tasks such as inspecting wings under a microscope, analyzing fossil material, and observing pinned specimens. The development process included the creation of digital assets and interaction design to ensure an accessible educational experience. Results: Diptera-VR is expected to bridge the gap between theoretical learning and scientific practice, increasing student engagement and improving knowledge retention. Additionally, it may contribute to the popularization of science in a context marked by a global environmental crisis and growing scientific denialism.
- Open Access
- 23 September 2025
- Published by Sociedade Brasileira de Computação
- p. 17-24
The Usability Engineering Lifecycle in Developing a Ludic Application about Scorpions for Children
Introduction: Given the increasing number of scorpion-related accidents, especially involving children, it is essential to develop educational strategies that promote awareness and prevention of such incidents. Objective: In this context, this article presents the practical application of the Usability Engineering Lifecycle in developing an educational ludic application aimed at children aged 5 to 10. The main goal is to create a playful and accessible tool that helps disseminate information about the risks and safety measures related to scorpions. Methodology: The methodology followed the classic stages of the Usability Engineering Lifecycle - Requirements Analysis, Design, Evaluation, and Development - carried out with the collaboration of experts in health, education, and technology, ensuring that the content and interface were appropriate for the target audience. Results: The results indicate that the developed ludic application has significant potential to promote children’s awareness, contributing to the prevention of scorpion accidents. The study also highlights the importance of integrating usability principles from the early stages of development. It emphasizes the relevance of participatory, user-centered approaches in the context of public health solutions.
- 23 September 2025
- Published by Sociedade Brasileira de Computação
- p. 17-24
Open AccessThe Usability Engineering Lifecycle in Developing a Ludic Application about Scorpions for Children
Introduction: Given the increasing number of scorpion-related accidents, especially involving children, it is essential to develop educational strategies that promote awareness and prevention of such incidents. Objective: In this context, this article presents the practical application of the Usability Engineering Lifecycle in developing an educational ludic application aimed at children aged 5 to 10. The main goal is to create a playful and accessible tool that helps disseminate information about the risks and safety measures related to scorpions. Methodology: The methodology followed the classic stages of the Usability Engineering Lifecycle - Requirements Analysis, Design, Evaluation, and Development - carried out with the collaboration of experts in health, education, and technology, ensuring that the content and interface were appropriate for the target audience. Results: The results indicate that the developed ludic application has significant potential to promote children’s awareness, contributing to the prevention of scorpion accidents. The study also highlights the importance of integrating usability principles from the early stages of development. It emphasizes the relevance of participatory, user-centered approaches in the context of public health solutions.
- Open Access
- 16 September 2025
- Published by IOS Press
- p. 135-148
Reinforcement Learning for Circular Sparsest Packing Problems
The sparsest packing problem emerges in manufacturing multi-hole extrusion dies to obtain small and precise products in the automotive, aviation, food, and medical industries. The goal is to maximize the minimum Euclidean distance between the objects and between the objects to the boundaries of the container. Additionally, this task might be subject to balancing constraints that determine that the deviation of gravity center of the system should stay within a threshold. We present a novel custom environment that encompasses the constraints present in this task. We experiment with the proposed environment using Proximal Policy Optimization to assess the applicability of reinforcement learning for the sparsest packing problem with circular objects in a circular container. Our results indicate that the proposed agent learns efficiently, demonstrating promising results in both finding feasible solutions and optimizing the placement of objects. Our approach not only shows the potential of reinforcement learning for solving the sparsest packing problem but also provides insights into its effectiveness in environments with complex spatial and balancing constraints.
- 16 September 2025
- Published by IOS Press
- p. 135-148
Open AccessReinforcement Learning for Circular Sparsest Packing Problems
The sparsest packing problem emerges in manufacturing multi-hole extrusion dies to obtain small and precise products in the automotive, aviation, food, and medical industries. The goal is to maximize the minimum Euclidean distance between the objects and between the objects to the boundaries of the container. Additionally, this task might be subject to balancing constraints that determine that the deviation of gravity center of the system should stay within a threshold. We present a novel custom environment that encompasses the constraints present in this task. We experiment with the proposed environment using Proximal Policy Optimization to assess the applicability of reinforcement learning for the sparsest packing problem with circular objects in a circular container. Our results indicate that the proposed agent learns efficiently, demonstrating promising results in both finding feasible solutions and optimizing the placement of objects. Our approach not only shows the potential of reinforcement learning for solving the sparsest packing problem but also provides insights into its effectiveness in environments with complex spatial and balancing constraints.