Abstract: Taxonomy-based image generation is used for item searching, and enhances the quality and personalization of search results. Prior interacted items are classified into a categorical taxonomy. A generative AI model can be used to classify the prior interacted items, by generating categories or assigning to existing categories. A set of prior interacted items is selected from one of the categories and provided to an image model that generates a photo-realistic image in response. The photo-realistic image includes item renderings that are rendered illustrations of items. An item search using a search engine can be performed based on the generated photo-realistic image. For instance, the photo-realistic image or a portion thereof could be provided as a search query using an image-based search or described to perform a text-based search. Search results are identified for the search query and are provided in response.

Abstract: Image-based searching can provide enhanced search results compared to text-based searches. Techniques for generating images that can be used by search engines include generating an optimized image-model prompt using a language model from a text-based input. The optimized image-model prompt includes a more literal description of an item described by the text-based input. The optimized image-model prompt is provided to a language model that generates a photo-realistic image of the item. A search engine uses the photo-realistic image to identify and return search results.

Abstract: An application server associated with an online marketplace may receive a set of images of an item, spatial information associated with each image of the set of images indicating a relative spatial location to the item for each image, and viewing direction information indicating a viewing direction relative to the item for each image. The application server may then generate a neural radiance field model of the item in a three dimensional space based on the set of images, the spatial information, and the viewing direction information, and may generate a set of spherical harmonics for the neural radiance field model based on predicting one or more spherical harmonic coefficients using the set of images, the spatial information, and the viewing direction information. The application server may then output a three dimensional model of the item generated based on the set of spherical harmonics.

Abstract: An application server associated with an online marketplace may receive a set of images of an item, spatial information associated with each image of the set of images indicating a relative spatial location to the item for each image, and viewing direction information indicating a viewing direction relative to the item for each image. The application server may then generate a neural radiance field model of the item in a three dimensional space based on the set of images, the spatial information, and the viewing direction information, and may generate a set of spherical harmonics for the neural radiance field model based on predicting one or more spherical harmonic coefficients using the set of images, the spatial information, and the viewing direction information. The application server may then output a three dimensional model of the item generated based on the set of spherical harmonics.

Abstract: Systems and methods for assessing pain in a subject are disclosed. The system can receive pain data indicative of pain in the subject, such as a patient pain drawing or a dermatome map. A processor circuit can use the pain data to generate a plurality of pain descriptors indicative of pain intensity, spatial extent of pain, or temporal pattern of pain. The processor can use the pain descriptors to produce a composite pain score that quantifies overall pain of the subject. The system can include an output unit that can produce a presentation of information about the pain in the subject including the composite pain score.