Raster2Seq generates labeled polygon sequences autoregressively from floorplan images via an anchor-guided decoder, claiming state-of-the-art results on Structure3D, CubiCasa5K, Raster2Graph and generalization to WAFFLE.
Nguyen, Yiwen Chen, Vikram V oleti, Varun Jam- pani, and Huaizu Jiang
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
SBM tokenizes building rooms via a sparse attribute-feature matrix and trains a Transformer for high-fidelity embeddings plus autoregressive layout generation, yielding better retrieval and fewer layout errors than baselines.
citing papers explorer
-
Raster2Seq: Polygon Sequence Generation for Floorplan Reconstruction
Raster2Seq generates labeled polygon sequences autoregressively from floorplan images via an anchor-guided decoder, claiming state-of-the-art results on Structure3D, CubiCasa5K, Raster2Graph and generalization to WAFFLE.
-
Tokenizing Buildings: A Transformer for Layout Synthesis
SBM tokenizes building rooms via a sparse attribute-feature matrix and trains a Transformer for high-fidelity embeddings plus autoregressive layout generation, yielding better retrieval and fewer layout errors than baselines.