Framework for Motorcycle Risk Assessment Using Onboard Panoramic Camera (Short Paper)

Authors Natchapon Jongwiriyanurak , Zichao Zeng , Meihui Wang , James Haworth , Garavig Tanaksaranond , Jan Boehm

Thumbnail PDF


  • Filesize: 2.83 MB
  • 7 pages

Document Identifiers

Author Details

Natchapon Jongwiriyanurak
  • Department of Civil, Environmental and Geomatic Engineering, University College London, UK
Zichao Zeng
  • Department of Civil, Environmental and Geomatic Engineering, University College London, UK
Meihui Wang
  • Department of Civil, Environmental and Geomatic Engineering, University College London, UK
James Haworth
  • Department of Civil, Environmental and Geomatic Engineering, University College London, UK
Garavig Tanaksaranond
  • Department of Survey Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, Thailand
Jan Boehm
  • Department of Civil, Environmental and Geomatic Engineering, University College London, UK

Cite AsGet BibTex

Natchapon Jongwiriyanurak, Zichao Zeng, Meihui Wang, James Haworth, Garavig Tanaksaranond, and Jan Boehm. Framework for Motorcycle Risk Assessment Using Onboard Panoramic Camera (Short Paper). In 12th International Conference on Geographic Information Science (GIScience 2023). Leibniz International Proceedings in Informatics (LIPIcs), Volume 277, pp. 44:1-44:7, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2023)


Traditional safety analysis methods based on historical crash data and simulation models have limitations in capturing real-world driving scenarios. In this experiment, panoramic videos recorded from a motorcyclist’s helmet in Bangkok, Thailand, were narrated using an image-to-text model and then put into a Large Language Model (LLM) to identify potential hazards and assess crash risks. The framework can assess static and moving objects with the potential for early warning and incident analysis. However, the limitations of the existing image-to-text model cause its inability to handle panoramic images effectively.

Subject Classification

ACM Subject Classification
  • Information systems → Geographic information systems
  • Computing methodologies → Scene understanding
  • Traffic incident risk
  • Large Language Model
  • Vision-Language Model


  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    PDF Downloads


  1. Gowri Asaithambi, Venkatesan Kanagaraj, and Tomer Toledo. Driving Behaviors: Models and Challenges for Non-Lane Based Mixed Traffic. Transportation in Developing Economies, 2(2):19, October 2016. URL:
  2. Jun Chen, Deyao Zhu, Kilichbek Haydarov, Xiang Li, and Mohamed Elhoseiny. Video ChatCaptioner: Towards Enriched Spatiotemporal Descriptions, April 2023. arXiv:2304.04227 [cs]. URL:
  3. Rupam Deb and Alan Wee-chung Liew. Missing Value Imputation for the Analysis of Incomplete Traffic Accident Data. In Xizhao Wang, Witold Pedrycz, Patrick Chan, and Qiang He, editors, Machine Learning and Cybernetics, volume 481, pages 275-286. Springer Berlin Heidelberg, Berlin, Heidelberg, 2014. Series Title: Communications in Computer and Information Science. URL:
  4. Nopadon Kronprasert, Chomphunut Sutheerakul, Thaned Satiennam, and Paramet Luathep. Intersection Safety Assessment Using Video-Based Traffic Conflict Analysis: The Case Study of Thailand. Sustainability, 13(22):12722, November 2021. URL:
  5. Gabriel Lanzaro, Tarek Sayed, and Rushdi Alsaleh. Can motorcyclist behavior in traffic conflicts be modeled? A deep reinforcement learning approach for motorcycle-pedestrian interactions. Transportmetrica B: Transport Dynamics, 10(1):396-420, December 2022. URL:
  6. Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models, May 2023. arXiv:2301.12597 [cs]. URL:
  7. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual Instruction Tuning, April 2023. arXiv:2304.08485 [cs]. URL:
  8. Jesus Perez-Martin, Benjamin Bustos, Silvio Jamil F. Guimarães, Ivan Sipiran, Jorge Pérez, and Grethel Coello Said. A Comprehensive Review of the Video-to-Text Problem, November 2021. arXiv:2103.14785 [cs]. URL:
  9. Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning Transferable Visual Models From Natural Language Supervision, February 2021. arXiv:2103.00020 [cs]. URL:
  10. E. Sanchez Castillo, D. Griffiths, and J. Boehm. SEMANTIC SEGMENTATION OF TERRESTRIAL LIDAR DATA USING CO-REGISTERED RGB DATA. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLIII-B2-2021:223-229, June 2021. URL:
  11. Chamroeun Se, Thanapong Champahom, Sajjakaj Jomnonkwao, and Vatanavongs Ratanavaraha. Motorcyclist injury severity analysis: a comparison of Artificial Neural Networks and random parameter model with heterogeneity in means and variances. International Journal of Injury Control and Safety Promotion, pages 1-16, June 2022. URL:
  12. Junke Wang, Dongdong Chen, Chong Luo, Xiyang Dai, Lu Yuan, Zuxuan Wu, and Yu-Gang Jiang. ChatVideo: A Tracklet-centric Multimodal and Versatile Video Understanding System, April 2023. arXiv:2304.14407 [cs]. URL:
  13. WHO. Global status report on road safety 2018. Technical Report 2, WHO, 2018. ISBN: 9789290496977 ISSN: 00142972 Publication Title: World Health Organization Volume: 3. URL:
  14. Ou Zheng. ChatGPT Is on the Horizon: Could a Large Language Model Be All We Need for Intelligent Transportation? Computation and Language, March 2023. URL: