Paper · ≈ 1 min read
NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction
Outgoing relations
- addresses → Novel View Synthesis (Task)
- evaluated_on → DTU (Benchmark)
- implemented_in → About Totoro97/NeuS (Repository)
- introduces → NeuS (Algorithm)
- part_of → 3D/4D Vision and Reconstruction (ResearchField)
- uses_metric → LPIPS (Metric)
- uses_metric → PSNR (Metric)
- uses_metric → SSIM (Metric)
Incoming relations
- mentioned_in → About Totoro97/NeuS (Repository)
Connected node types
- Metric: 3
- Repository: 2
- Algorithm: 1
- Benchmark: 1
- ResearchField: 1
- Task: 1
Mentions in the corpus
- part_of Evidence: Meanwhile, recent neural methods for novel view synthesis, such as NeRF…
- mentioned_in Claim: Meanwhile, recent neural methods for novel view synthesis, such as NeRF and its variants, use v…
- part_of Evidence: We present a novel neural surface reconstruction method, called NeuS, f…
- part_of Evidence: In NeuS, we propose to represent a surface as the zero-level set of a s…
- part_of Evidence: Experiments on the DTU dataset and the BlendedMVS dataset show that Neu…
- mentioned_in About Totoro97/NeuS
- 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-2106-10689/abstract.md