ScVLM: Enhancing Vision-Language Model for Safety-Critical Event Understanding
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929749999222784 |
|---|---|
| author | Shi, Liang Jiang, Boyu Zeng, Tong Guo, Feng |
| author_facet | Shi, Liang Jiang, Boyu Zeng, Tong Guo, Feng |
| contents | Accurately identifying, understanding and describing traffic safety-critical events (SCEs), including crashes, tire strikes, and near-crashes, is crucial for advanced driver assistance systems, automated driving systems, and traffic safety. As SCEs are rare events, most general vision-language models (VLMs) have not been trained sufficiently to link SCE videos and narratives, which could lead to hallucinations and missing key safety characteristics. Here, we introduce ScVLM, a novel hybrid methodology that integrates supervised and contrastive learning techniques to classify the severity and types of SCEs, as well as to generate narrative descriptions of SCEs. This approach utilizes classification to enhance VLMs' comprehension of driving videos and improve the rationality of event descriptions. The proposed approach is trained on and evaluated by more than 8,600 SCEs from the Second Strategic Highway Research Program Naturalistic Driving Study dataset, the largest publicly accessible driving dataset with videos and SCE annotations. The results demonstrate the superiority of the proposed approach in generating contextually accurate event descriptions and mitigating VLM hallucinations. The code will be available at https://github.com/datadrivenwheels/ScVLM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_00982 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ScVLM: Enhancing Vision-Language Model for Safety-Critical Event Understanding Shi, Liang Jiang, Boyu Zeng, Tong Guo, Feng Computer Vision and Pattern Recognition Accurately identifying, understanding and describing traffic safety-critical events (SCEs), including crashes, tire strikes, and near-crashes, is crucial for advanced driver assistance systems, automated driving systems, and traffic safety. As SCEs are rare events, most general vision-language models (VLMs) have not been trained sufficiently to link SCE videos and narratives, which could lead to hallucinations and missing key safety characteristics. Here, we introduce ScVLM, a novel hybrid methodology that integrates supervised and contrastive learning techniques to classify the severity and types of SCEs, as well as to generate narrative descriptions of SCEs. This approach utilizes classification to enhance VLMs' comprehension of driving videos and improve the rationality of event descriptions. The proposed approach is trained on and evaluated by more than 8,600 SCEs from the Second Strategic Highway Research Program Naturalistic Driving Study dataset, the largest publicly accessible driving dataset with videos and SCE annotations. The results demonstrate the superiority of the proposed approach in generating contextually accurate event descriptions and mitigating VLM hallucinations. The code will be available at https://github.com/datadrivenwheels/ScVLM. |
| title | ScVLM: Enhancing Vision-Language Model for Safety-Critical Event Understanding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.00982 |