Paper · ≈ 1 min read
LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation
Outgoing relations
- part_of → 3D/4D Vision and Reconstruction (ResearchField)
- uses → Volumetric Rendering (MethodologicalConcept)
- uses_metric → LPIPS (Metric)
- uses_metric → PSNR (Metric)
- uses_metric → SSIM (Metric)
Connected node types
- Metric: 3
- MethodologicalConcept: 1
- ResearchField: 1
Mentions in the corpus
- part_of Evidence: In this paper, we introduce Large Multi-View Gaussian Model (LGM), a no…
- part_of Evidence: Our key insights are two-fold: 1) 3D Representation: We propose multi-v…
- part_of Evidence: 2) 3D Backbone: We present an asymmetric U-Net as a high-throughput bac…
- part_of Evidence: 3D content creation has achieved significant progress in terms of both…
- part_of Evidence: Two design choices make this practical at high resolution: the multi-vi…
- mentioned_in Claim: Two design choices make this practical at high resolution: the multi-view Gaussian features are…
- summarizes Project Pulse
- summarizes Field Overview — 3D/4D Vision and Reconstruction
Cross-references in raw data
Source provenance
examples/demo-corpus/data/research/papers/arxiv-2402-05054/abstract.md