Abstract: Adaptive road surface detection and safety feature activation includes receiving an image captured by a camera, the image comprising a road surface ahead of a vehicle. It further includes providing the image to an image-based road surface condition prediction model that is trained at least in part using physically-sensed traction data. It further includes pre-emptively activating one or more safety-related features of the vehicle based at least in part on an anticipated condition of the road surface predicted by the image-based road surface condition prediction model.

Abstract: Determining an absolute depth estimate of a vehicle object in a 2D image captured by a camera includes receiving the 2D image captured by the camera. It further includes determining, at least in part by using a segmentation model, a plurality of pixels corresponding to the vehicle object in the 2D image. It furhter includes using a prediction model to determine, for each pixel in the plurality of pixels determined at least in part by using the segmentation model, a correspondng camera invariant distance value comprising a predicted height of a cross-section of the vehicle object that is visible within a given pixel. It further includes determining the absolute depth estimate of the vehicle object in the 2D image based on: a focal length of the camera that captured the 2D image; and an aggregation of the predicted heights determined for each of the pixels in the plurality of pixels.

Abstract: Generative image synthesis conditioned on label prompts includes receiving a label prompt comprising a mapping of labels to portions of an image to be synthesized. At least some of the labels in the label prompt correspond to an object to be included in the image to be synthesized. It further includes, based on the label prompt, generating, using a generative model, a synthesized image corresponding to the label prompt.