Abstract: Provided are machine learning models that generate geometry-free neural scene representations through efficient object-centric novel-view synthesis. In particular, one example aspect of the present disclosure provides a novel framework in which an encoder model (e.g., an encoder transformer network) processes one or more RGB images (with or without pose) to produce a fully latent scene representation that can be passed to a decoder model (e.g., a decoder transformer network). Given one or more target poses, the decoder model can synthesize images in a single forward pass. In some example implementations, because transformers are used rather than convolutional or MLP networks, the encoder can learn an attention model that extracts enough 3D information about a scene from a small set of images to render novel views with correct projections, parallax, occlusions, and even semantics, without explicit geometry.

Abstract: Systems and methods for providing agricultural operation(s) for a crop, in manner consistent with a predicted quality of the crop are described. One example computer-implemented method includes accessing, by a computing device, data associated with a crop in a target field, where the data includes planting data for the crop and weather data for the target field, and correlating the weather data to each of a plurality of growth stages for the crop in the target field. The method then includes, prior to a harvest of the crop from the target field, determining, by the computing device, a quality of the crop in the target field, based on the correlated weather data and a quality model specific to a type of the crop, and directing at least one agricultural operation consistent with the determined quality of the crop in the target field.