Paper · ≈ 1 min read
DeepV2D: Video to Depth with Differentiable Structure from Motion
Outgoing relations
- addresses → Visual SLAM (ResearchTopic)
- introduces → DeepV2D (Algorithm)
- part_of → 3D/4D Vision and Reconstruction (ResearchField)
- uses_metric → LPIPS (Metric)
- uses_metric → PSNR (Metric)
- uses_metric → SSIM (Metric)
Connected node types
- Metric: 3
- Algorithm: 1
- ResearchField: 1
- ResearchTopic: 1
Mentions in the corpus
- part_of Evidence: --- type: Paper arxiv: "1812.04605" arxiv_url: https://arxiv.org/abs/18…
- mentioned_in Claim: --- type: Paper arxiv: "1812.04605" arxiv_url: https://arxiv.org/abs/1812.04605 title: "DeepV2D…
- part_of Evidence: We propose DeepV2D, an end-to-end deep learning architecture for predic…
- summarizes Project Pulse
- synthesizes Topic — Visual SLAM
- summarizes Topic — Visual SLAM
- summarizes Field Overview — 3D/4D Vision and Reconstruction
Cross-references in raw data
Source provenance
examples/demo-corpus/data/research/papers/arxiv-1812-04605/abstract.md