FlowSSC: Universal Generative Monocular Semantic Scene Completion via One-Step Latent Diffusion
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_ | 1866908780055232512 |
|---|---|
| author | Xi, Zichen Chen, Hao-Xiang Xue, Nan Yan, Hongyu Feng, Qi-Yuan Kara, Levent Burak Jorge, Joaquim Xu, Qun-Ce |
| author_facet | Xi, Zichen Chen, Hao-Xiang Xue, Nan Yan, Hongyu Feng, Qi-Yuan Kara, Levent Burak Jorge, Joaquim Xu, Qun-Ce |
| contents | Semantic Scene Completion (SSC) from monocular RGB images is a fundamental yet challenging task due to the inherent ambiguity of inferring occluded 3D geometry from a single view. While feed-forward methods have made progress, they often struggle to generate plausible details in occluded regions and preserve the fundamental spatial relationships of objects. Such accurate generative reasoning capability for the entire 3D space is critical in real-world applications. In this paper, we present FlowSSC, the first generative framework applied directly to monocular semantic scene completion. FlowSSC treats the SSC task as a conditional generation problem and can seamlessly integrate with existing feed-forward SSC methods to significantly boost their performance. To achieve real-time inference without compromising quality, we introduce Shortcut Flow-matching that operates in a compact triplane latent space. Unlike standard diffusion models that require hundreds of steps, our method utilizes a shortcut mechanism to achieve high-fidelity generation in a single step, enabling practical deployment in autonomous systems. Extensive experiments on SemanticKITTI demonstrate that FlowSSC achieves state-of-the-art performance, significantly outperforming existing baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_15250 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FlowSSC: Universal Generative Monocular Semantic Scene Completion via One-Step Latent Diffusion Xi, Zichen Chen, Hao-Xiang Xue, Nan Yan, Hongyu Feng, Qi-Yuan Kara, Levent Burak Jorge, Joaquim Xu, Qun-Ce Computer Vision and Pattern Recognition Robotics Semantic Scene Completion (SSC) from monocular RGB images is a fundamental yet challenging task due to the inherent ambiguity of inferring occluded 3D geometry from a single view. While feed-forward methods have made progress, they often struggle to generate plausible details in occluded regions and preserve the fundamental spatial relationships of objects. Such accurate generative reasoning capability for the entire 3D space is critical in real-world applications. In this paper, we present FlowSSC, the first generative framework applied directly to monocular semantic scene completion. FlowSSC treats the SSC task as a conditional generation problem and can seamlessly integrate with existing feed-forward SSC methods to significantly boost their performance. To achieve real-time inference without compromising quality, we introduce Shortcut Flow-matching that operates in a compact triplane latent space. Unlike standard diffusion models that require hundreds of steps, our method utilizes a shortcut mechanism to achieve high-fidelity generation in a single step, enabling practical deployment in autonomous systems. Extensive experiments on SemanticKITTI demonstrate that FlowSSC achieves state-of-the-art performance, significantly outperforming existing baselines. |
| title | FlowSSC: Universal Generative Monocular Semantic Scene Completion via One-Step Latent Diffusion |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2601.15250 |