Computer Vision Research Intern – Neural Reconstruction & Generation
About the Internship
Join our team as a research intern working on 3D reconstruction and generative AI for sensor simulation. You’ll work closely with engineers and researchers on real-world sensor data, and turn research ideas into working prototypes. Strong results may lead to publications at top venues.
What You’ll Do
· Research and implement 3D reconstruction and neural rendering methods, such as 3D Gaussian Splatting, for large-scale, dynamic driving scenes.
· Experiment with video diffusion models for visual generation tasks, such as novel view synthesis.
· Work with camera, LiDAR and other sensor data.
· Run experiments, define realism metrics, and report results clearly.
· Contribute to research code and, where possible, publications.
Required Qualifications
· Currently pursuing an MS or PhD in Computer Vision, Machine Learning, Graphics, Robotics or a related field.
· Solid 3D vision fundamentals: camera models, multi-view geometry and depth estimation.
· Hands-on research or project experience with 3D Gaussian Splatting or other neural rendering methods, or video diffusion models.
· Strong Python and PyTorch skills.
Preferred Qualifications
· Experience with LiDAR or multi-sensor data.
· Publications at top venues (CVPR, ICCV, ECCV, NeurIPS, SIGGRAPH), or strong open-source work.
· C++ or CUDA.
· Camera and video data processing experience.
