# Remote-Sensing City Layout Extraction with MLLM

arXiv: https://arxiv.org/abs/2608.16484 · HTML: https://mlsanity.com/paper/2608.16484/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 13 references that are in the map.
- First posted on arXiv 17 Aug 2026.

## Builds on (most cited first, 12 of 12)

- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018)
- [DOTA: A Large-scale Dataset for Object Detection in Aerial Images](https://mlsanity.com/paper/1711.10398/) (2017)
- [ViperGPT: Visual Inference via Python Execution for Reasoning](https://mlsanity.com/paper/2303.08128/) (2023)
- [Visual Programming: Compositional visual reasoning without training](https://mlsanity.com/paper/2211.11559/) (2022)
- [LayoutGPT: Compositional Visual Planning and Generation with Large Language Models](https://mlsanity.com/paper/2305.15393/) (2023)
- [Holodeck: Language Guided Generation of 3D Embodied AI Environments](https://mlsanity.com/paper/2312.09067/) (2023)
- [iSAID: A Large-scale Dataset for Instance Segmentation in Aerial Images](https://mlsanity.com/paper/1905.12886/) (2019)
- [3D-GPT: Procedural 3D Modeling with Large Language Models](https://mlsanity.com/paper/2310.12945/) (2023)
- [CityJSON: a compact and easy-to-use encoding of the CityGML data model](https://mlsanity.com/paper/1902.09155/) (2019)
- [IRSAMap:Towards Large-Scale, High-Resolution Land Cover Map Vectorization](https://mlsanity.com/paper/2508.16272/) (2025)
- [Code-as-Room: Generating 3D Rooms from Top-Down View Images via Agentic Code Synthesis](https://mlsanity.com/paper/2605.18451/) (2026)

## Abstract (arXiv)

Remote-sensing systems usually describe urban content with detection boxes, semantic masks, or vector boundaries. Such outputs locate classes and support image-plane scoring, yet they do not by themselves constitute an executable layout that retains object identities, typed relations, topology, and regeneration rules. Code-as-City instead casts urban-layout extraction from a single top-down image as constrained code generation with a multimodal large language model (MLLM). An image model first produces an aligned five-class semantic layout prior. Three ordered MLLM passes use the image and this prior to recover roads, land-cover regions and relations, and buildings. Deterministic normalization converts the accumulated records into a city graph and a restricted layout program. Executing the program creates a renderable 3D city layout and an orthographic semantic projection over shared geometry. The projection admits pixel-level comparison with remote-sensing masks, while named objects, relations, and editing operations remain available for synchronized regeneration of both views. Evaluated on the 100 scenes of CityLayout-100, the complete framework obtains 41.1% mean intersection-over-union and 48.3% global intersection-over-union. This result provides quantitative evidence that visual observations can be translated into inspectable, editable city code with coupled planar and 3D outputs.