Search papers, labs, and topics across Lattice.
The paper introduces DM-OSVP++, a novel one-shot view planning method for active object reconstruction that leverages 3D diffusion models as priors. By conditioning on initial multi-view images, the method generates an approximate 3D object model using the diffusion prior, which is then used to inform the selection of informative viewpoints. The key result is a system that effectively integrates geometric and textural information from the diffusion-based object model into the view planning process, leading to improved reconstruction performance in both simulation and real-world experiments.
3D diffusion models can be harnessed to generate object priors for one-shot view planning, enabling robots to efficiently reconstruct objects by focusing on complex regions.
Active object reconstruction is crucial for many robotic applications. A key aspect in these scenarios is generating object-specific view configurations to obtain informative measurements for reconstruction. One-shot view planning enables efficient data collection by predicting all views at once, eliminating the need for time-consuming online replanning. Our primary insight is to leverage the generative power of 3D diffusion models as valuable prior information. By conditioning on initial multi-view images, we exploit the priors from the 3D diffusion model to generate an approximate object model, serving as the foundation for our view planning. Our novel approach integrates the geometric and textural distributions of the object model into the view planning process, generating views that focus on the complex parts of the object to be reconstructed. We validate the proposed active object reconstruction system through both simulation and real-world experiments, demonstrating the effectiveness of using 3D diffusion priors for one-shot view planning.
