Search papers, labs, and topics across Lattice.
GaussLite introduces a task-conditioned approach to 3D Gaussian Splatting (3DGS) that optimizes representation density based on specific robotic tasks, enhancing efficiency in real-time mapping. By utilizing a one-shot LLM parser to identify relevant objects and generate per-pixel relevance masks, the system dynamically allocates computational resources to the most pertinent aspects of the scene. In evaluations, GaussLite significantly outperformed existing methods, achieving an average increase of 2.72 dB in ROI PSNR on the Replica Dataset and demonstrating superior performance in real-world applications.
Task-driven mapping with GaussLite boosts 3D reconstruction quality by over 2.7 dB while optimizing compute resources for robotic tasks.
Existing 3D Gaussian Splatting (3DGS) systems distribute representation capacity uniformly across a scene, ignoring the fact that many downstream robotic tasks engage only a fraction of the reconstructed geometry. This causes valuable onboard compute to be allocated towards optimizing irrelevant parts of the scene, either limiting online capacity or under-optimizing the most relevant parts of the scene. We introduce GaussLite, a task-driven 3DGS mapping system that conditions its representation density on a natural-language task specification. Given a posed RGB-D stream and a task such as "prepare to pick up the object on the desk," GaussLite uses a one-shot LLM parser to extract target and anchor objects, which are grounded per-frame by an open-vocabulary detector and segmented to produce per-pixel relevance masks in real time. The mapper allocates seeding density, gradient flow and scaling by task relevance. At matched Gaussian budget and real-time mapping at 4 Hz on resource-constrained hardware, GaussLite outperforms baselines on ROI PSNR on the Replica Dataset by an average +2.72 dB and on a real-hardware demonstration in indoor and outdoor settings by +2.23 dB. We further show that two task-specialized agents' maps can be fused into a single shared map via per-voxel voting on active-optimization counts in real time, outperforming concatenation by +3.42 dB while only sharing an average 7.08% of the map.
