Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.02164 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912566553346048 |
|---|---|
| author | Zhang, Xinshen Fu, Tongxi Zheng, Xu |
| author_facet | Zhang, Xinshen Fu, Tongxi Zheng, Xu |
| contents | Omnidirectional scene understanding is vital for various downstream applications, such as embodied AI, autonomous driving, and immersive environments, yet remains challenging due to geometric distortion and complex spatial relations in 360° imagery. Existing omnidirectional methods achieve scene understanding within a single frame while neglecting cross-frame correlated panoramas. To bridge this gap, we introduce \textbf{CFpano}, the \textbf{first} benchmark dataset dedicated to cross-frame correlated panoramas visual question answering in the holistic 360° scenes. CFpano consists of over 2700 images together with over 8000 question-answer pairs, and the question types include both multiple choice and open-ended VQA. Building upon our CFpano, we further present \methodname, a multi-modal large language model (MLLM) fine-tuned with Group Relative Policy Optimization (GRPO) and a set of tailored reward functions for robust and consistent reasoning with cross-frame correlated panoramas. Benchmark experiments with existing MLLMs are conducted with our CFpano. The experimental results demonstrate that \methodname achieves state-of-the-art performance across both multiple-choice and open-ended VQA tasks, outperforming strong baselines on all major reasoning categories (\textbf{+5.37\%} in overall performance). Our analyses validate the effectiveness of GRPO and establish a new benchmark for panoramic scene understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_02164 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Omnidirectional Spatial Modeling from Correlated Panoramas Zhang, Xinshen Fu, Tongxi Zheng, Xu Computer Vision and Pattern Recognition Omnidirectional scene understanding is vital for various downstream applications, such as embodied AI, autonomous driving, and immersive environments, yet remains challenging due to geometric distortion and complex spatial relations in 360° imagery. Existing omnidirectional methods achieve scene understanding within a single frame while neglecting cross-frame correlated panoramas. To bridge this gap, we introduce \textbf{CFpano}, the \textbf{first} benchmark dataset dedicated to cross-frame correlated panoramas visual question answering in the holistic 360° scenes. CFpano consists of over 2700 images together with over 8000 question-answer pairs, and the question types include both multiple choice and open-ended VQA. Building upon our CFpano, we further present \methodname, a multi-modal large language model (MLLM) fine-tuned with Group Relative Policy Optimization (GRPO) and a set of tailored reward functions for robust and consistent reasoning with cross-frame correlated panoramas. Benchmark experiments with existing MLLMs are conducted with our CFpano. The experimental results demonstrate that \methodname achieves state-of-the-art performance across both multiple-choice and open-ended VQA tasks, outperforming strong baselines on all major reasoning categories (\textbf{+5.37\%} in overall performance). Our analyses validate the effectiveness of GRPO and establish a new benchmark for panoramic scene understanding. |
| title | Omnidirectional Spatial Modeling from Correlated Panoramas |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.02164 |