Split Liability Assessment in Car Accident using 3D Convolutional Neural Network

Abstract In a car accident, negligence is evaluated through process known as split liability assessment. This assessment involves reconstructing the accident scenario based on information gathered from sources such dashcam footage. The final determination of made by simulating contained in video. Therefore, cases for should be classified affecting degree. While deep learning has recently been spotlight video recognition using short clips, no research conducted to extract meaningful long videos, which are necessary To address this issue, we propose new task analyzing videos stacking important predicted 3D CNNs model. We demonstrate feasibility our approach proposing method