eVGGT: An Efficient Geometry-Aware Vision Encoder for Visuomotor Policies
This work introduces eVGGT, a lightweight geometry-aware vision encoder distilled from a larger model, designed to improve robotic imitation learning by incorporating 3D spatial reasoning. By replacing standard 2D vision encoders with eVGGT's latent space in frameworks like ACT and DP, the method boosts success rates by up to 6.5% while being significantly faster and smaller.