Open to opportunities

Shiva User

@shivauser10

Message

I build computer vision pipelines for 3D reconstruction, pose estimation, and medical imaging.

India

Message

What I'm looking for

I'm looking to build production and research-focused computer vision systems, especially in 3D reconstruction, pose estimation, multimodal understanding, and medical imaging, while growing through challenging engineering work.

I built synthetic occlusion-aware 3D human pose benchmarks at UiT The Arctic University of Norway, combining SMPL-X, AMASS motion sequences, Unity data generation, and 3D Gaussian Splatting across 8+ environments.

At Humantics, I designed and shipped a production medical computer vision pipeline for image segmentation, DICOM processing, and automated delivery across clinical imaging data. I automated Orthanc-based ingestion and segmentation workflows with MongoDB, Supabase, AWS Lambda, FastAPI, and GCP, and built a mobile-first DICOM viewer using Capacitor and OHIF.

I also reimplemented and extended COSMOS for sparse-view 3D Gaussian Splatting and built a multimodal fashion-video understanding pipeline using YOLOv8, CLIP, FAISS, Whisper, and spaCy.

Experience

Work history, roles, and key accomplishments

Computer Vision Research Intern

UiT The Arctic University of Norway

Jun 2025 - Aug 2025 (2 months)

Built a synthetic occlusion-aware 3D human pose benchmark combining SMPL-X models with 3D Gaussian Splatting scenes. Developed a Unity data-generation pipeline producing 1000+ RGB frames with ground-truth joints and occlusion labels.

Gaussian Splats Unity Amass Synthetic Data Generation Benchmarking