2025/11/14 by de Araujo, Camila Machado, Borges, Egon P. B. S., Grangeiro, Ricardo Marcelo Canteiro +1
#62H35 #68U10 #68W10 #Computer Vision and Pattern Recognition (cs.CV) #Distributed #FOS: Computer and information sciences #I.4.3 #I.4.6 #I.4.7 #I.5.3 #I.5.4 #Parallel #and Cluster Computing (cs.DC)
paper · doi:10.48550/arxiv.2511.11890
High-resolution volumetric imaging techniques, such as X-ray tomography and advanced microscopy, generate increasingly large datasets that challenge existing tools for efficient processing, segmentation, and interactive exploration. This work introduces new capabilities to Annotat3D through Harpia, a new CUDA-based processing library designed to support scalable, interactive segmentation workflows for large 3D datasets in high-performance computing (HPC) and remote-access environments. Harpia features strict memory control, native chunked execution, and a suite of GPU-accelerated filtering, annotation, and quantification tools, enabling reliable operation on datasets exceeding single-GPU memory capacity. Experimental results demonstrate significant improvements in processing speed, memory efficiency, and scalability compared to widely used frameworks such as NVIDIA cuCIM and scikit-image. The system's interactive, human-in-the-loop interface, combined with efficient GPU resource management, makes it particularly suitable for collaborative scientific imaging workflows in shared HPC infrastructures.