Abstract: The subject application relates to a server, terminal and non-transitory computer-readable medium. The server for handling streaming data for a live streaming, comprising one or a plurality of processors, wherein the one or plurality of processors execute a machine-readable instruction to perform: recording the streaming data for the live streaming; storing the streaming data as archive contents with first identifier; receiving interaction information during the live streaming; storing the interaction information as contexts with second identifier, transmitting the archive contents with first identifier to a first user terminal; and transmitting the contexts to the first user terminal according to the first identifier and the second identifier. According to the subject application, the archive contents may be more immersive and the user experience may be enhanced.

Abstract: A traffic condition prediction system and a traffic condition prediction method are disclosed. The method includes: determining a center of circle in a surveillance image of a target traffic scene; determining a first circle based on the center of circle and a first radius; extracting a plurality of first feature points along the circumference of the first circle according to a first preset sampling frequency; generating a scene feature of the target traffic scene at least based on the first feature points; determining whether the scene feature and a scene feature of another traffic scene are similar; and when determining that they are similar, predicting traffic condition of the target traffic scene through a prediction model used for predicting traffic condition of the other traffic scene. The scene feature of the target traffic scene and that of the other traffic scene are generated in a same way.

Abstract: A computing device divides a training image into a plurality of training blocks, and the training image includes a training object. The computing device calculates, for each of the training blocks, a correct confidence score of the training object covering the training block according to an image-marking data and a confidence-score-translating function, and the image-marking data includes a piece of location information of the training object in the training image. Then, the computing device trains a deep-learning model with the training image, the correct confidence scores and the image-marking data to generate the object-detecting model.

Abstract: A traffic condition prediction system and a traffic condition prediction method are disclosed. The method includes: determining a center of circle in a surveillance image of a target traffic scene; determining a first circle based on the center of circle and a first radius; extracting a plurality of first feature points along the circumference of the first circle according to a first preset sampling frequency; generating a scene feature of the target traffic scene at least based on the first feature points; determining whether the scene feature and a scene feature of another traffic scene are similar; and when determining that they are similar, predicting traffic condition of the target traffic scene through a prediction model used for predicting traffic condition of the other traffic scene. The scene feature of the target traffic scene and that of the other traffic scene are generated in a same way.

Abstract: A computing device divides a training image into a plurality of training blocks, and the training image includes a training object. The computing device calculates, for each of the training blocks, a correct confidence score of the training object covering the training block according to an image-marking data and a confidence-score-translating function, and the image-marking data includes a piece of location information of the training object in the training image. Then, the computing device trains a deep-learning model with the training image, the correct confidence scores and the image-marking data to generate the object-detecting model.