ROOT: VLM based System for Indoor Scene Understanding and Beyond
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929603667296256 |
|---|---|
| author | Wang, Yonghui Chen, Shi-Yong Zhou, Zhenxing Li, Siyi Li, Haoran Zhou, Wengang Li, Houqiang |
| author_facet | Wang, Yonghui Chen, Shi-Yong Zhou, Zhenxing Li, Siyi Li, Haoran Zhou, Wengang Li, Houqiang |
| contents | Recently, Vision Language Models (VLMs) have experienced significant advancements, yet these models still face challenges in spatial hierarchical reasoning within indoor scenes. In this study, we introduce ROOT, a VLM-based system designed to enhance the analysis of indoor scenes. Specifically, we first develop an iterative object perception algorithm using GPT-4V to detect object entities within indoor scenes. This is followed by employing vision foundation models to acquire additional meta-information about the scene, such as bounding boxes. Building on this foundational data, we propose a specialized VLM, SceneVLM, which is capable of generating spatial hierarchical scene graphs and providing distance information for objects within indoor environments. This information enhances our understanding of the spatial arrangement of indoor scenes. To train our SceneVLM, we collect over 610,000 images from various public indoor datasets and implement a scene data generation pipeline with a semi-automated technique to establish relationships and estimate distances among indoor objects. By utilizing this enriched data, we conduct various training recipes and finish SceneVLM. Our experiments demonstrate that \rootname facilitates indoor scene understanding and proves effective in diverse downstream applications, such as 3D scene generation and embodied AI. The code will be released at \url{https://github.com/harrytea/ROOT}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_15714 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ROOT: VLM based System for Indoor Scene Understanding and Beyond Wang, Yonghui Chen, Shi-Yong Zhou, Zhenxing Li, Siyi Li, Haoran Zhou, Wengang Li, Houqiang Computer Vision and Pattern Recognition Recently, Vision Language Models (VLMs) have experienced significant advancements, yet these models still face challenges in spatial hierarchical reasoning within indoor scenes. In this study, we introduce ROOT, a VLM-based system designed to enhance the analysis of indoor scenes. Specifically, we first develop an iterative object perception algorithm using GPT-4V to detect object entities within indoor scenes. This is followed by employing vision foundation models to acquire additional meta-information about the scene, such as bounding boxes. Building on this foundational data, we propose a specialized VLM, SceneVLM, which is capable of generating spatial hierarchical scene graphs and providing distance information for objects within indoor environments. This information enhances our understanding of the spatial arrangement of indoor scenes. To train our SceneVLM, we collect over 610,000 images from various public indoor datasets and implement a scene data generation pipeline with a semi-automated technique to establish relationships and estimate distances among indoor objects. By utilizing this enriched data, we conduct various training recipes and finish SceneVLM. Our experiments demonstrate that \rootname facilitates indoor scene understanding and proves effective in diverse downstream applications, such as 3D scene generation and embodied AI. The code will be released at \url{https://github.com/harrytea/ROOT}. |
| title | ROOT: VLM based System for Indoor Scene Understanding and Beyond |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2411.15714 |