Search papers, labs, and topics across Lattice.
This paper introduces Local View Synthesis (LVS), a framework that reuses previously rendered RGB-D images to synthesize new views in interactive 3D scene exploration, avoiding costly full scene re-rendering. By employing geometric warping guided by relative camera pose and a lightweight multiscale network for residual prediction, LVS effectively recovers newly exposed content and corrects artifacts. The method achieves a 0.72 dB PSNR improvement over pure warping on GS-render and demonstrates low query latency, enabling responsive scene exploration.
Interactive 3D scene exploration can now achieve responsive view updates by intelligently reusing rendered images, bypassing expensive full scene re-rendering.
Interactive scene exploration requires frequent view updates, although small camera motions preserve much of the visible content. Conventional 3D Gaussian Splatting nevertheless renders each target view, leaving this image overlap unexploited. Reusing rendered images offers an alternative. Geometric warping alone cannot recover newly exposed content and remains sensitive to depth errors. We propose a per-scene framework that replaces repeated scene rendering for nearby views with relative-pose-guided RGB-D image reuse. Geometric warping uses depth and relative pose to transport source content, while a lightweight multiscale network predicts RGB residuals to correct artifacts and infer missing appearance. Cached source features further reduce repeated computation. On GS-render, residual refinement improves PSNR by 0.72~dB over pure warping; evaluations on captured and rendered scenes demonstrate low query latency. This separation of scene rendering from local view updates supports responsive scene exploration, with potential applications in augmented and virtual reality.
