Abstract: The subject technology provides a framework for learning neural scene representations directly from images, without three-dimensional (3D) supervision, by a machine-learning model. In the disclosed systems and methods, 3D structure can be imposed by ensuring that the learned representation transforms like a real 3D scene. For example, a loss function can be provided which enforces equivariance of the scene representation with respect to 3D rotations. Because naive tensor rotations may not be used to define models that are equivariant with respect to 3D rotations, a new operation called an invertible shear rotation is disclosed, which has the desired equivariance property. In some implementations, the model can be used to generate a 3D representation, such as mesh, of an object from an image of the object.

Abstract: A method and process for simultaneously searching one or more internet auctions, augmented with automatic seller ratings expressing risk-of-injury to a buyer, with the steps of: A search parameter entry client, An artificial intelligence-based analysis engine that generates ratings of sellers expressing risk-of-injury to a buyer, An easy to understand ‘rating’ display automatically presented to the buyer, A server system that accepts requests for rating sellers and returns the desired ratings, A web browser or instant messenger component or web page that monitors a buyer's current URL and which interacts with the server system to return and display the seller's ‘rating’ when the buyer is browsing an online auction listing, A web browser or instant messenger component or web page that interacts with the server system to allow a search of the online auction service platform which returns a sorted list of rated sellers.

Abstract: A method for enabling risk management for sellers of items on Internet auction sites with the steps of: Automatically detect new bids on auctions, Evaluate bidders on the basis of their past performance using an artifical intelligence system, Maintain on a central server rating information for bidders, Maintain on the central server a seller's preferences for dealing with bidders in each rating category, Based upon the rating act on the auction by: canceling bid, notifying seller, or accepting the bid, Based upon the bidder's country of registration notify the seller, Allow sellers to see all auctions with bidders on one screen, Allow sellers to see a list of other auctions where their bidders have placed bids, Automate auction fee refund processing, and Allow sellers to see a list of all pending and recently completed auction refund requests, allow anonymous reprisal-free bidder feedback.