Research map / 3D vision / Human pose estimation
SliceMatch: Geometry-Guided Aggregation for Cross-View Pose Estimation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use geometry-guided, ground-slice-conditioned descriptors for pose estimation | VIGOR cross-area pose estimation · VIGOR cross-area, unaligned | median location error (m): 9.55 -> 5.64 | Helped |
| Use geometry-guided, ground-slice-conditioned descriptors for pose estimation | VIGOR cross-area pose estimation · VIGOR cross-area, unaligned | median orientation error (degrees): 29.97 -> 5.18 | Helped |
| Use SliceMatch rather than LM | KITTI same-area pose estimation with a 20° orientation prior · KITTI same-area, 20° orientation prior | median location error (m): 11.42 -> 4.39 | Helped |
| Use SliceMatch rather than LM | KITTI cross-area pose estimation with a 20° orientation prior · KITTI cross-area, 20° orientation prior | mean location error (m): 12.58 -> 13.50 | Hurt |
| Use SliceMatch rather than LM | KITTI cross-area pose estimation with a 20° orientation prior · KITTI cross-area, 20° orientation prior | median location error (m): 12.11 -> 9.77 | Helped |
| Increase the number of slices from 1 to 16 (ablation) | SliceMatch · VIGOR tuning split | mean location error (m): 12.73 -> 7.60 | Helped |
| Increase the number of slices from 16 to 32 (ablation) | SliceMatch · VIGOR tuning split | mean location error (m): 7.60 -> 8.14 | Hurt |
| Add cross-view attention (ablation) | SliceMatch · VIGOR tuning split | median orientation error (degrees): 12.32 -> 9.22 | Helped |
| Use the reweighted contrastive loss instead of original infoNCE (ablation) | SliceMatch · VIGOR tuning split | mean location error (m): 8.08 -> 7.60 | Helped |
| Increase the number of slices from 8 to 16 (ablation) | SliceMatch · KITTI same-area, 20° orientation prior | mean location error (m): 8.74 -> 7.96 | Helped |
| Increase the number of slices from 16 to 32 (ablation) | SliceMatch · KITTI same-area, 20° orientation prior | mean location error (m): 7.96 -> 8.03 | Hurt |
| Train with small-HFoV ground images | SliceMatch · VIGOR same-area, limited-HFoV test images | location error | Helped |
Closest papers by meaning
- FG^2: Fine-Grained Cross-View Localization by Fine-Grained Feature Matching (CVPR 2025 · similarity 0.89)
- View From Above: Orthogonal-View aware Cross-view Localization (CVPR 2024 · similarity 0.89)
- BevSplat: Resolving Height Ambiguity via Feature-Based Gaussian Primitives for Weakly-Supervised Cross-View Localization (NeurIPS 2025 · similarity 0.86)
- From Coarse to Fine: A Matching and Alignment Framework for Unsupervised Cross-View Geo-Localization (AAAI 2025 · similarity 0.86)
- Seeing Across Skies and Streets: Feedforward 3D Reconstruction from Satellite, Drone, and Ground Images (NeurIPS 2026 · similarity 0.85)
- Aligning Geometric Spatial Layout in Cross-View Geo-Localization via Feature Recombination (AAAI 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.