Sparse-view 4D Gaussian Splatting for robot manipulation scenes

Developed a sparse view 4D Gaussian Splatting reconstruction algorithm for use in real-world robot tasks. Conducted real-world experiments that demonstrated improved geometric and photometric quality, along with stronger temporal adaptability, validating the system’s effectiveness beyond simulation. The algorithm cuts the number of input views needed for reconstruction from ∼20 to ∼4. The resulting controllable 4D scene representation was further integrated into an adversarial iterative Diffusion Policy learning framework, where scene variations in object configurations and appearance were used to expose policy failure modes and guide targeted retraining.

During student researcher at Robot Vision and Learning Lab