Research map / 3D vision / 3D generation
URoPE: Universal Relative Position Embedding across Geometric Spaces
Yichen Xie, Depu Meng, Yihan Hu, Chensheng Peng, Quentin HERAU, Masayoshi TOMIZUKA, Wei Zhan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| replace Plucker ray with URoPE | LVSM · RealEstate10k | PSNR: 23.95 -> 26.02 | Improved |
| URoPE vs P-RoPE and RayRoPE | LVSM · Objaverse | PSNR: 24.88 / 24.96 -> 25.09 | Improved |
| add URoPE to PETR | PETR · nuScenes | NDS: 34.9 -> 37.3 | Improved |
| add URoPE to StreamPETR | StreamPETR · nuScenes | NDS: 47.6 -> 50.6 | Improved |
| add URoPE to UniMatch stereo | UniMatch · Scenes11 | Abs Rel: 0.065 -> 0.049 | Improved |
| URoPE vs P-RoPE in stereo | UniMatch · Scenes11 | Abs Rel: 0.049 -> 0.049 | No clear effect |
| add 2D-RoPE-only relative embedding to Plucker (ablation) | LVSM · RealEstate10k | PSNR: 23.95 -> 24.96 | Improved |
| use head-wise instead of channel-wise anchor split (ablation) | URoPE · RealEstate10k | PSNR: 25.47 -> 25.94 | Improved |
| use 4 fixed anchors vs 1 (ablation) | URoPE · RealEstate10k | PSNR: 25.37 -> 26.01 | Improved |
| vary anchor count from 4 to 16 (ablation) | URoPE · RealEstate10k | PSNR: 26.01 -> 25.85 | No clear effect |
| fixed anchors instead of learned depth module (ablation) | URoPE · RealEstate10k | PSNR: 25.57 -> 26.01 | Improved |
| add Plucker absolute embedding to URoPE (ablation) | URoPE · RealEstate10k | PSNR: 25.89 vs 25.85 without | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- RayRoPE — RayRoPE: Projective Ray Positional Encoding for Multi-view Attention (ECCV 2026) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Cameras as Relative Positional Encoding (NeurIPS 2025 · similarity 0.90)
- ComRoPE: Scalable and Robust Rotary Position Embedding Parameterized by Trainable Commuting Angle Matrices (CVPR 2025 · similarity 0.86)
- DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers (NeurIPS 2026 · similarity 0.86)
- Maximizing the Position Embedding for Vision Transformers with Global Average Pooling (AAAI 2025 · similarity 0.85)
- Circle-RoPE: Cone-like Decoupled Rotary Positional Embedding for Vision-Language Models (ICML 2026 · similarity 0.85)
- GTA: A Geometry-Aware Attention Mechanism for Multi-View Transformers (ICLR 2024 · similarity 0.84)
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.
