Abstract: Systems, methods, and storage media for operating a machine learning system for identifying an object in a photograph are disclosed. The system is configured to prepare a plurality of data files for training the machine learning system by associating a label with each of the plurality of data files, splitting the plurality of data files into different data sets, including full, training, testing, and validation data sets, creating a final machine learning model based on metrics and/or artifacts associated with training on the full data set, and deploying the final machine learning model to a machine learning system endpoint. The system is further configured to identify at least one object in a user uploaded photograph based on invoking a first or a second trained model, the first or second trained model associated with the final machine learning model.

Abstract: A system is described herein comprising an application running on a processor of a mobile device, the application receiving a query image, wherein the application is communicatively coupled with one or more applications running on at least one processor of at least one remote server. The system includes the application providing the query image to the one or more applications, the one or more applications processing the query image to identify a query type corresponding to the query image, wherein the query type comprises a plurality of species. The system includes using a query type recognition engine corresponding to the query type to process the query image, the processing the query image including identifying at least one species corresponding to the query image. The system includes the one or more applications providing information of the at least one species to the application, wherein the application displays the information.

Abstract: A system is described that comprises an application configured to use an object detection model to detect and locate an image category across image frames of image data in real time, the detecting and locating including visualizing the location of the image category in a highlighted view across the image frames using an electronic display, the detecting and locating including capturing a frame of the image data as an image for image recognition. The application is configured to provide the image to one or more applications running on a remote server, the one or more applications configured to process the image to identify a species of a plant appearing in the image, the one or more applications configured to provide an identification of the species to the application, the application configured to receive an instruction to post the image and the species identification.

Abstract: A system is described herein comprising an application running on a processor of a mobile device, the application receiving a query image, wherein the application is communicatively coupled with one or more applications running on at least one processor of at least one remote server. The system includes the application providing the query image to the one or more applications, the one or more applications processing the query image to identify a query type corresponding to the query image, wherein the query type comprises a plurality of species. The system includes using a query type recognition engine corresponding to the query type to process the query image, the processing the query image including identifying at least one species corresponding to the query image. The system includes the one or more applications providing information of the at least one species to the application, wherein the application displays the information.

Abstract: Innovations are provided for encoding and/or decoding video and/or image content using transform coefficient level gradual updating. Transform coefficient level gradual updating can be applied by encoding (or decoding) different subsets of the transform coefficients for the blocks, macroblocks, or other coding unit for each of a sequence of pictures. For example, a first subset of the transform coefficients of the blocks of a first picture can be encoded with the first picture, a second subset of the transform coefficients of the blocks of a second picture can be encoded with the second picture, and so on. A decoder can reconstruct pictures with increasing quality by receiving additional subsets of the transform coefficients.

Abstract: Innovations are provided for encoding and/or decoding video and/or image content using transform coefficient level gradual updating. Transform coefficient level gradual updating can be applied by encoding (or decoding) different subsets of the transform coefficients for the blocks, macroblocks, or other coding unit for each of a sequence of pictures. For example, a first subset of the transform coefficients of the blocks of a first picture can be encoded with the first picture, a second subset of the transform coefficients of the blocks of a second picture can be encoded with the second picture, and so on. A decoder can reconstruct pictures with increasing quality by receiving additional subsets of the transform coefficients.