Scenario Extraction from Drive Recorder Footage for Traffic Scene Retrieval
交通シーン検索のための車載カメラ映像からのシナリオ抽出技術
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20265207
- Paper/Info type
- Proceedings (Spring)
No.49-26
- Pages
- 1-6(Total 6 p)
- Date of publication
- May 2026
- Publisher
- JSAE
- Language
- Japanese
- Event
- 2026 JSAE Annual Congress (Spring)
Detailed Information
| Author(J) | 1) 露木 雅文, 2) 西田 哲哉, 3) 冨田 民則, 4) 新 吉高 |
|---|---|
| Author(E) | 1) Masafumi Tsuyuki, 2) Tetsuya Nishida, 3) Taminori Tomita, 4) Yoshitaka Atarashi |
| Affiliation(J) | 1) 日立製作所, 2) Astemo, 3) Astemo, 4) Astemo |
| Affiliation(E) | 1) Hitachi, 2) Astemo, 3) Astemo, 4) Astemo |
| Abstract(J) | 先進運転支援システムの開発効率化には映像検索が有効であるが，多様な検索ニーズへの網羅的対応は困難である．本研究では運転シミュレータが任意の交通シーンを再現できる点に着目，そのシナリオ定義を活用した検索を提案する．特に，単眼カメラ映像と位置情報のみからシナリオを抽出し，検索可能にする方式を示す． Translation |
| Abstract(E) | Enhancing the development efficiency of Advanced Driver Assistance Systems (ADAS) requires robust video retrieval methods. However, comprehensively addressing diverse retrieval requirements remains a significant challenge. This study introduces a scenario-based retrieval approach that leverages the capability of driving simulators to reproduce arbitrary traffic scenes. The proposed method extracts scenario definitions exclusively from monocular camera footage and positional data, enabling structured and searchable representations. By facilitating efficient access to relevant driving situations, this approach supports systematic evaluation and accelerates the development of advanced driver assistance technologies. |
