Hi 👋 I'm Juan Pablo Suárez

Computer Vision Engineer | Remote Sensing AI | Geospatial Intelligence

Hi! I'm Juan Pablo, an Electronic Engineer from the Universidad Industrial de Santander (Colombia) with a strong focus on Computer Vision, Artificial Intelligence, and Deep Learning. I'm passionate about developing intelligent vision systems capable of understanding visual information through object detection, semantic and instance segmentation, image classification, and multimodal data analysis, transforming research into practical solutions for real-world applications.

Currently, I work on developing and evaluating AI solutions for remote sensing, precision agriculture, and environmental monitoring, building computer vision pipelines from satellite imagery, drone imagery, LiDAR point clouds, and IoT data. My work includes dataset preparation, annotation workflows, quality assurance, model training, performance evaluation, and the development of robust deep learning models for geospatial applications. I have also participated in Computer Vision quality assurance projects, auditing large-scale image annotations and validating AI models through quantitative performance metrics such as Precision, Recall, F1-score, and Confusion Matrix analysis.

My research interests include Computer Vision, Deep Learning, Foundation Models, Geospatial AI, Medical Imaging, and Multimodal Learning, with a particular interest in developing AI systems that bridge academic research and real-world impact.

I'm always interested in collaborating on research projects, summer research programs, research internships, and opportunities to pursue M.Sc. or Ph.D. studies in collaboration with universities, research laboratories, and industry partners.

🌴 ForestSAM A Novel Integration of DeepForest and SAM2 for Oil Palm Crown Segmentation in Aerial Imagery A hybrid Computer Vision framework that integrates DeepForest and Segment Anything Model 2 (SAM2) for accurate oil palm crown segmentation from aerial imagery, improving precision agriculture and remote sensing applications. 📄 Paper: https://link.springer.com/chapter/10.1007/978-3-032-23161-1_3 💻 Code: Coming Soon

🫀 Pulmonary Arterial Segmentation Deep Learning-Based Pulmonary Arterial Segmentation in Computed Tomography Images Deep learning approach for automatic pulmonary artery segmentation in CT images, supporting accurate and efficient medical image analysis. 📄 Paper: https://ieeexplore.ieee.org/document/10637810

• 🎩 Card Magic
• 🪙 Numismatics
• 🌲 Nature & Outdoor Activities