We propose a new label adaptation technique that first extracts internal states of vehicles from the underlying driving simulator, and then refines labels.

Developing a computer vision-based algorithm for identifying dangerous vehicles requires a large amount of labeled accident data, which is difficult to collect ...

Red bounding boxes de- note dangerous vehicles while green ones denote safe ones. For all the other vehicles (in an accident scene) that are not going to crash ...

Code for the paper "Crash To Not Crash: Learn to Identify Dangerous Vehicles Using a Simulator" presented in AAAI 2019 (Oral) - gnsrla12/crash_to_not_crash.

Jan 27, 2019 · Via real-data experiments, we show that our dangerous vehicle classifier can reduce the missed detection rate by at least 18.5% compared with ...

Oct 22, 2024 · Developing a computer vision-based algorithm for identifying dangerous vehicles requires a large amount of labeled accident data, ...

Jul 17, 2019 · This work proposes a new label adaptation technique that first extracts internal states of vehicles from the underlying driving simulator, ...

Dive into the research topics of 'Crash to not crash: Learn to identify dangerous vehicles using a simulator'. Together they form a unique fingerprint.

Jul 27, 2023 · A deep-learning-based vehicle collision prediction system was developed using data collected from a video game. The proposed system outperforms ...

Mar 1, 2024 · Crash to not crash: Learn to identify dangerous vehicles using a simulator. ... Vehicle accident and traffic classification using deep ...