Generating 3D House Wireframes with Semantics
This work addresses the need for detailed and semantically meaningful 3D wireframes in architectural design or computer graphics, representing an incremental advance in generative modeling for 3D structures.
The paper tackles the problem of generating 3D house wireframes with semantic enrichment by using an autoregressive model with a unified wire-based representation, resulting in improved accuracy and semantic fidelity compared to existing generative models.
We present a new approach for generating 3D house wireframes with semantic enrichment using an autoregressive model. Unlike conventional generative models that independently process vertices, edges, and faces, our approach employs a unified wire-based representation for improved coherence in learning 3D wireframe structures. By re-ordering wire sequences based on semantic meanings, we facilitate seamless semantic integration during sequence generation. Our two-phase technique merges a graph-based autoencoder with a transformer-based decoder to learn latent geometric tokens and generate semantic-aware wireframes. Through iterative prediction and decoding during inference, our model produces detailed wireframes that can be easily segmented into distinct components, such as walls, roofs, and rooms, reflecting the semantic essence of the shape. Empirical results on a comprehensive house dataset validate the superior accuracy, novelty, and semantic fidelity of our model compared to existing generative models. More results and details can be found on https://vcc.tech/research/2024/3DWire.