{"slug": "trace-gs-on-policy-trajectory-distillation-with-privileged-geometric-for-sparse", "title": "TRACE-GS: On-Policy Trajectory Distillation with Privileged Geometric Conditioning for Sparse-View 3DGS Restoration", "summary": "TRACE-GS, a new framework from arXiv (paper 2608.10286v1), performs on-policy trajectory distillation with privileged geometric conditioning to improve sparse-view 3D Gaussian Splatting (3DGS) restoration. It uses a teacher model with richer geometry from additional training views to guide the student model along its own rollout, achieving consistent gains and strong generalization across datasets and sparse-view settings.", "body_md": "arXiv:2608.10286v1 Announce Type: new\nAbstract: We present TRACE-GS, an on-policy trajectory distillation framework that leverages privileged geometric conditioning at training time, thereby adapting a diffusion prior to sparse-view 3D Gaussian Splatting (3DGS) restoration. Rather than pursuing increasingly sophisticated restoration architectures, we identify a more fundamental limitation shared by existing diffusion-based approaches: supervision at independently noised states does not cover those reached during inference. In sparse-view 3DGS, under-constrained geometry biases denoising from the outset, and the resulting deviations compound along the rollout. TRACE-GS instead performs on-policy trajectory distillation: a teacher conditioned on richer geometry from additional training views supplies targets along the sparse-view student's own rollout, aligning denoising directions and cross-view responses at each visited state. This training-only geometry places TRACE-GS in the learning using privileged information (LUPI) setting. At deployment, only the sparse-view student is retained, and its restored renderings serve as pseudo-observations for 3DGS refinement. To the best of our knowledge, TRACE-GS is the first to derive on-policy supervision from privileged geometry for sparse-view 3DGS restoration, achieving consistent gains and strong generalization across datasets and sparse-view settings.", "url": "https://wpnews.pro/news/trace-gs-on-policy-trajectory-distillation-with-privileged-geometric-for-sparse", "canonical_source": "https://arxiv.org/abs/2608.10286", "published_at": "2026-08-12 04:00:00+00:00", "updated_at": "2026-08-12 04:11:27.677109+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision", "generative-ai"], "entities": ["TRACE-GS", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/trace-gs-on-policy-trajectory-distillation-with-privileged-geometric-for-sparse", "markdown": "https://wpnews.pro/news/trace-gs-on-policy-trajectory-distillation-with-privileged-geometric-for-sparse.md", "text": "https://wpnews.pro/news/trace-gs-on-policy-trajectory-distillation-with-privileged-geometric-for-sparse.txt", "jsonld": "https://wpnews.pro/news/trace-gs-on-policy-trajectory-distillation-with-privileged-geometric-for-sparse.jsonld"}}