FOFernando Santos Osorio

Fernando Santos Osorio

*This is an algorithmically generated author record based on publicly available information.

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


Fernando Santos Osorio has a total of 156 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Evaluating Monocular Depth Estimation on Embedded Platforms for Autonomous Navigation

    by Wilbur Naike Chiuyari Veramendi, Fernando Santos Osório

    Deep learning-based monocular depth estimation has achieved significant advancements on urban benchmarks, but its embedded application remains limited by efficiency constraints. Vision Transformers (ViTs) and Foundation Models (FMs) show promising zero-shot generalization capabilities, yet their adaptation to resource-constrained hardware requires careful study. In this work, we investigate the development of the DepthAnything model on an NVIDIA Jetson Orin, analyzing the trade-off between accuracy and inference speed for different backbones (ViT-S, ViT-B, and ViT-L). We report quantitative metrics including AbsRel, δ1, RMSE, and FPS on the KITTI dataset, along with qualitative results. Our experiments show that the ViT-S backbone offers the best balance of accuracy and real-time performance (44 FPS), whereas ViT-B suffers from degradation and ViT-L exhibits significant instability due to optimization artifacts. These findings highlight the viability of compact backbones for embedded visual perception and suggest future optimizations, such as quantization-aware training and pruning, in larger architectures.

  • Open Access

    Efficient Pothole Segmentation Using Quantization and Pruning for ADAS

    by William D'ABruzzo Martins, Fernando Santos Osório, Diego Renan Bruno

    Real-time semantic segmentation is critical for Advanced Driver-Assistance Systems (ADAS), but deploying deep learning models on resource-constrained hardware remains a challenge. This work presents an efficient pipeline for pothole segmentation using the lightweight PP-LiteSeg model, evaluating and comparing two key optimization techniques: post-training quantization and network pruning. Results demonstrate that 8-bit integer (INT8) quantization is highly effective, achieving a 3.54x speedup in GPU compute time with only a minimal, well-controlled loss in segmentation accuracy. While network pruning also reduces model size and latency, it comes at a more pronounced cost to performance. In conclusion, while quantization offers a superior trade-off between speed and accuracy, network pruning remains a valuable strategy for applications where memory compression is the primary objective.

  • Open Access

    “Photoguide” para telemedicina – Aplicativo móvel para captura de fotos com padrão técnico por pacientes e compartilhamento com seus médicos

    by Leonardo Montenegro Matos Albuquerque, Elaine Kawano Horibe, Sheila Souza Sales, Angelica Cardoso Martins, Fernanda Feijoeiro, Fernando Santos Osorio

    Resumo: O uso de fotografias auxilia a comunicação entre médicos e pacientes na telemedicina. Atualmente, o uso de aplicativos de mensagens para envio de fotos traz problemas como a falta de obtenção de consentimento, a falta de segurança e de privacidade, a ausência de registro adequado em prontuário médico e a má qualidade das imagens registradas. Desenvolver um aplicativo móvel para captura de fotos com padrão técnico por pacientes e compartilhamento com seus médicos para uso em telemedicina. Busca de anterioridade e revisão de literatura nortearam o desenvolvimento do aplicativo pela técnica de design thinking. Foram entrevistados médicos e pacientes para definir os problemas a serem solucionados. O aplicativo desenvolvido foi submetido a testes de usabilidade para aprimoramento da versão final que foi entregue. Na busca de anterioridade, não foram encontrados aplicativos voltados para a padronização de fotos pelo próprio paciente. Foram entrevistados 10 médicos: 80% recebiam fotos várias vezes por semana, 90% consideravam importantes as imagens na tomada de decisões, 100% usavam WhatsApp (Meta Platforms, Inc.) e, apesar de 100% considerarem importante obter consentimento, 90% raramente o faziam. A qualidade das fotos foi considerada importante por 90% deles. Dos 11 pacientes entrevistados, 64% enviavam fotos aos médicos com frequência, e 90% consideravam as imagens muito importantes. O Whatsapp foi o aplicativo mais utilizado. Falta de segurança, baixa qualidade, enquadramento e foco foram os problemas definidos a serem solucionados. Brainstorming entre pesquisadores, médicos e programadores guiou a prototipação, a validação e o desenvolvimento do aplicativo. O aplicativo Photoguide foi desenvolvido e entregue.

  • Drowsiness Detection Using Vital Sign Sensors and Deep Learning on Smartwatches

    by Vitor Augusto da Rosa Pereira, Rafael Alceste Berri, Fernando Santos Osório

  • Open Access

    Integrating Modular Pipelines with End-to-End Learning: A Hybrid Approach for Robust and Reliable Autonomous Driving Systems

    by Luis Alberto Rosero, Iago Pachêco Gomes, Júnior Anderson Rodrigues da Silva, Carlos André Przewodowski, Denis Fernando Wolf, Fernando Santos Osório

    Autonomous driving navigation relies on diverse approaches, each with advantages and limitations depending on various factors. For HD maps, modular systems excel, while end-to-end methods dominate mapless scenarios. However, few leverage the strengths of both. This paper innovates by proposing a hybrid architecture that seamlessly integrates modular perception and control modules with data-driven path planning. This innovative design leverages the strengths of both approaches, enabling a clear understanding and debugging of individual components while simultaneously harnessing the learning power of end-to-end approaches. Our proposed architecture achieved first and second place in the 2023 CARLA Autonomous Driving Challenge’s SENSORS and MAP tracks, respectively. These results demonstrate the architecture’s effectiveness in both map-based and mapless navigation. We achieved a driving score of 41.56 and the highest route completion of 86.03 in the MAP track of the CARLA Challenge leaderboard 1, and driving scores of 35.36 and 1.23 in the CARLA Challenge SENSOR track with route completions of 85.01 and 9.55, for, respectively, leaderboard 1 and 2. The results of leaderboard 2 raised the hybrid architecture to the first position, winning the edition of the 2023 CARLA Autonomous Driving Competition.


  • Open Access

    Global Localization using OpenStreetMap and Elevation Offsets

    by André Przewodowski, Fernando Santos Osório, Valdir Grassi Junior

    Localization is a critical component in autonomous vehicle navigation stacks. While GNSS-only localization cannot be fully reliable and available all the time, localization based on 3D high-definition (HD) maps have to be robust to world changes, which is still a challenging issue. Added to that, in general, HD maps are expensive and difficult to construct and maintain.In this paper, we propose a particle filter-based 2D global pose estimation method that can use the crowdsourced OpenStreetMap (OSM) API, a digital surface map, or both. The main contributions of the proposed approach are: that it is lightweight, does not require the vehicle to map the environment, does not require a GPU (can be used with low-power computing resources), is agnostic to the odometry source, and achieved relatively low position and orientation errors for this localization modality using the KITTI dataset sequences. The proposed method's implementation is open source and is available with the experimental results on our GitHub page.


  • Open Access

    Integrating Modular Pipelines with End-to-End Learning: A Hybrid Approach for Robust and Reliable Autonomous Driving Systems

    by Luis Alberto Rosero, Iago Pachêco Gomes, Júnior Anderson Rodrigues Da Silva, Carlos André Braile Przewodowski Filho, Denis Fernando Wolf, Fernando Santos Osório

    Autonomous driving navigation relies on diverse approaches, each with advantages and limitations depending on various factors. For HD maps, modular systems excel, while end-to-end methods dominate mapless scenarios. However, few leverage the strengths of both. This paper innovates by proposing a hybrid architecture that seamlessly integrates modular perception and control modules with data-driven path planning. This innovative design leverages the strengths of both approaches, enabling a clear understanding and debugging of individual components while simultaneously harnessing the learning power of end-to-end approaches. Our proposed architecture achieved 1st and 2nd place in the 2023 CARLA Autonomous Driving Challenge’s SENSORS and MAP tracks, respectively. These results demonstrate the architecture’s effectiveness in both map-based and mapless navigation.


  • Open Access

    Lexical Analysis of Automatic Transcriptions Using Speech-to-Text Services: A Statistically Evaluated Case Study

    by Venilton Falvojr, Anderson Marcolino, Diego Bruno, Catherine Martins Falvo, Fernando Santos Osório, Ellen Barbosa

    This paper introduces Speech2Learning, an innovative architecture designed to leverage Speech-To-Text (STT) technology to enhance the accessibility of Learning Objects (LOs). Stemming from a recognized gap in prior Systematic Mapping, the primary objective of this architecture is to simplify the development of flexible educational solutions. In a collaborative endeavor with Brazilian EdTech DIO, we instantiated Speech2Learning as a Proof of Concept (PoC) to subtitle video lessons on their e-learning platform. This PoC was essential to obtain valuable insights for a more comprehensive Case Study. Therefore, we performed a lexical similarity analysis on the automatic transcriptions generated by leading STT providers in Portuguese, English and Spanish. Finally, we carried out a rigorous Statistical Analysis to evaluate the quantitative data from the Case Study. Our findings highlight the potential of Speech2Learning to promote the accessibility of LOs, as well as the relevance of continued research to increase the accuracy of STT services.