Methods, systems, and computer program products for detecting vehicles 221a 221c in low light conditions. Cameras 204 are used to obtain red, green, blue (RGB) images of the environment around a vehicle. RGB images are converted 213 to LAB images 233. The “A” channel is filtered to extract contours from LAB images 214. The contours are filtered based on their shapes/sizes to reduce false positives from contours unlikely to correspond to vehicles. A neural network 217 classifies an object as a vehicle or non-vehicle based the contours 237. Vehicles can be detected at night as well as in other low light conditions using their head lights and tail (rear) lights, enabling autonomous vehicles to better detect other vehicles in their environment. Vehicle detections can be facilitated using a combination of virtual data, deep learning, and computer vision. The contour may be sent along with range data from a LIDAR sensor 206 to the neural network.