Research map / 3D vision / 3D generation
OmniCity: Omnipotent City Understanding With Multi-Level and Multi-View Images
Weijia Li, Yawen Lai, Linning Xu, Yuanbo Xiangli, Jinhua Yu, Conghui He, Gui-Song Xia, Dahua Lin
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 |
|---|---|---|---|
| Use small rather than medium or large off-nadir view angle | Mask R-CNN building footprint segmentation · OmniCity satellite images | AP: 29.7 vs 23.7 vs 18.9 | Improved |
| Use DORN instead of SARPN | Satellite building height estimation · OmniCity small, medium, and large view angles | MSE: 870.34 -> 670.52; 694.17 -> 628.06; 823.01 -> 730.67 | Improved |
| Use medium rather than small or large view angle | SARPN and DORN height estimation · OmniCity satellite images | MAE: SARPN: 16.18, 13.75, 15.32; DORN: 12.71, 12.24, 13.40 | Improved |
| Use mono-view rather than panorama images | Street-level building instance segmentation · OmniCity matched collection sites | AP: 61.6 -> 68.3 | Improved |
| Use mono-view rather than panorama images | Street-level land use segmentation · OmniCity matched collection sites | AP: 20.6 -> 23.9 | Improved |
| Use hybrid task cascade rather than Mask R-CNN | Building footprint segmentation · OmniCity satellite, small view angle | AP: 29.7 -> 32.3 | Improved |
| Use hybrid task cascade rather than Mask R-CNN | Fine-grained land use segmentation · OmniCity panorama images | AP: 26.0 -> 27.2 | Improved |
| Use building instance segmentation rather than fine-grained land use segmentation | Mask R-CNN · OmniCity panorama images | AP: 26.0 -> 66.7 | Improved |
Closest papers by meaning
- SG-BEV: Satellite-Guided BEV Fusion for Cross-View Semantic Segmentation (CVPR 2024 · similarity 0.85)
- UrbanFeel:A Comprehensive Benchmark for Temporal and Perceptual Understanding of City Scenes through Human Perspective (ICLR 2026 · similarity 0.84)
- GeoFidelity-Bench: Evaluating Block-Conditioned Geographic Fidelity of Street-View Generation (NeurIPS 2026 · similarity 0.83)
- UrbanDataLayer: A Unified Data Pipeline for Urban Science (NeurIPS 2024 · similarity 0.83)
- UV-SAM: Adapting Segment Anything Model for Urban Village Identification (AAAI 2024 · similarity 0.83)
- DynamicVL: Benchmarking Multimodal Large Language Models for Dynamic City Understanding (NeurIPS 2025 · similarity 0.83)
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.
