About this role
, Netherlands Worker Type: Interns (Fixed Term) (Trainee) Posting Start Date: August 5, 2026 Job Description:
Objective The objective of this internship is to evaluate the quality of CAD models generated from images and videos through an existing internal reconstruction pipeline and to assess the effort required to prepare these models for meshing and subsequent engineering analyses. Scope of Work Phase 1: Evaluation of Reconstructed CAD Models The student will:
• Support the testing and usage of 3D reconstruction algorithms based on COLMAP, MVSFormer++, Trimesh, glomap, and Poisson reconstruction.
• Compare the generated CAD models against the original objects represented in the source images and videos.
• Assess the geometric fidelity and completeness of the reconstructed models.
• Identify reconstruction artefacts, missing features, distortions, and other quality issues.
• Define and apply suitable qualitative and quantitative criteria for model evaluation in collaboration with the Computer Vision team. Phase 2: Model Cleanup and Meshing Readiness Assessment The student will:
• Analyse the generated CAD geometries to determine their suitability for meshing operations with tools such as Ansys SpaceClaim or StarCCM (not exhaustive list)
• Identify common issues requiring cleanup, such as:
• Holes and gaps in surfaces
• Non-manifold geometries
• Excessive surface noise
• Redundant or disconnected features
• Geometric inconsistencies
• Evaluate the level of manual intervention required to obtain mesh-ready geometries with canonical cases
• Document the cleanup workflow and estimate the effort associated with different classes of models. Phase 3 (Optional Extension) Depending on the findings from Phases 1 and 2 and the available internship duration, the scope may be extended to investigate improvements in the upstream data-processing workflow. Potential activities include:
• Assessing the impact of image and video quality on reconstruction performance.
• Evaluating data acquisition best practices (camera positioning, lighting conditions, coverage, resolution, etc.).
• Investigating preprocessing methods for images and videos.
• Identifying opportunities to improve the robustness and accuracy of the reconstruction pipeline.
• Developing recommendations for data collection and processing guidelines. Expected Deliverables
• Evaluation methodology for assessing reconstruction quality.
• Benchmark dataset and documented assessment results.
• Analysis of typical reconstruction defects and cleanup requirements.
• Guidelines for preparing reconstructed models for meshing.
• Recommendations for improving the overall reconstruction workflow.
• Final presentation and internship report summarizing findings and conclusions.
-
DISCLAIMER: Please note: We occasionally amend or withdraw Shell jobs and reserve the right to do so at any time, including prior to the advertised closing date. Before applying, you are advised to read our data protection policy. This policy describes the processing that may be associated with your personal data and informs you that your personal data may be transferred to Shell/Shell Group companies around the world. The Shell Group and its approved recruitment consultants will never ask you for a fee to process or consider your application for a career with Shell. Anyone who demands such a fee is not an authorised Shell representative and you are strongly advised to refuse any such demand. Shell is an Equal Opportunity Employer.