PlaneCycle: Training-Free 2D-to-3D Lifting of Foundation Models Without Adapters

PlaneCycle introduces a training-free method to convert 2D AI foundation models to 3D without requiring retraining or architectural changes. The technique enables pretrained 2D models like DINOv3 to process 3D volumetric data by cyclically distributing spatial aggregation across orthogonal planes, achieving competitive performance on 3D classification and segmentation tasks.

• →PlaneCycle allows 2D foundation models to process 3D data without retraining, adapters, or architectural modifications
• →The method introduces zero additional parameters and works with arbitrary 2D neural networks
• →Testing with DINOv3 models showed competitive performance on nine 3D benchmarks including classification and segmentation tasks
• →Under linear probing, lifted models outperformed slice-wise 2D baselines and strong 3D counterparts
• →With full fine-tuning, PlaneCycle matches standard 3D architectures while preserving pretrained knowledge