PSGS: Text-driven Panorama Sliding Scene Generation via Gaussian Splatting
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911412921565184 |
|---|---|
| author | Zhang, Xin Chen, Shen Zhou, Jiale Li, Lei |
| author_facet | Zhang, Xin Chen, Shen Zhou, Jiale Li, Lei |
| contents | Generating realistic 3D scenes from text is crucial for immersive applications like VR, AR, and gaming. While text-driven approaches promise efficiency, existing methods suffer from limited 3D-text data and inconsistent multi-view stitching, resulting in overly simplistic scenes. To address this, we propose PSGS, a two-stage framework for high-fidelity panoramic scene generation. First, a novel two-layer optimization architecture generates semantically coherent panoramas: a layout reasoning layer parses text into structured spatial relationships, while a self-optimization layer refines visual details via iterative MLLM feedback. Second, our panorama sliding mechanism initializes globally consistent 3D Gaussian Splatting point clouds by strategically sampling overlapping perspectives. By incorporating depth and semantic coherence losses during training, we greatly improve the quality and detail fidelity of rendered scenes. Our experiments demonstrate that PSGS outperforms existing methods in panorama generation and produces more appealing 3D scenes, offering a robust solution for scalable immersive content creation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_00463 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PSGS: Text-driven Panorama Sliding Scene Generation via Gaussian Splatting Zhang, Xin Chen, Shen Zhou, Jiale Li, Lei Computer Vision and Pattern Recognition Generating realistic 3D scenes from text is crucial for immersive applications like VR, AR, and gaming. While text-driven approaches promise efficiency, existing methods suffer from limited 3D-text data and inconsistent multi-view stitching, resulting in overly simplistic scenes. To address this, we propose PSGS, a two-stage framework for high-fidelity panoramic scene generation. First, a novel two-layer optimization architecture generates semantically coherent panoramas: a layout reasoning layer parses text into structured spatial relationships, while a self-optimization layer refines visual details via iterative MLLM feedback. Second, our panorama sliding mechanism initializes globally consistent 3D Gaussian Splatting point clouds by strategically sampling overlapping perspectives. By incorporating depth and semantic coherence losses during training, we greatly improve the quality and detail fidelity of rendered scenes. Our experiments demonstrate that PSGS outperforms existing methods in panorama generation and produces more appealing 3D scenes, offering a robust solution for scalable immersive content creation. |
| title | PSGS: Text-driven Panorama Sliding Scene Generation via Gaussian Splatting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.00463 |