Abstract: An image feature extraction method for a 360° image includes the following steps: projecting the 360° image onto a cube model to generate an image stack including a plurality of images having a link relationship; using the image stack as an input of a neural network, wherein when operation layers of the neural network performs padding operation on one of the plurality of images, the link relationship between the plurality of adjacent images is used such that the padded portion at the image boundary is filled with the data of neighboring images in order to retain the characteristics of the boundary portion of the image; and by the arithmetic operation of the neural network of such layers with the padded feature map, an image feature map is generated.

Abstract: The disclosure proposes a method and a system for 360-degree video playback. The method is applicable to a video playback system and includes at least the following steps. A current frame of a 360-degree video having a sequence of frames is received. Candidate objects in the current frame are detected, and a main object is selected from the candidate objects by using a selector recurrent neural network (RNN) model based on information of the candidate objects in the current frame. A viewing angle corresponding to the current frame is computed by using a regressor RNN model based on the main object in the current frame. The current frame is displayed on the screen according to the viewing angle.

Abstract: The disclosure proposes a method and a system for 360-degree video playback. The method is applicable to a video playback system and includes at least the following steps. A current frame of a 360-degree video having a sequence of frames is received. Candidate objects in the current frame are detected, and a main object is selected from the candidate objects by using a selector recurrent neural network (RNN) model based on information of the candidate objects in the current frame. A viewing angle corresponding to the current frame is computed by using a regressor RNN model based on the main object in the current frame. The current frame is displayed on the screen according to the viewing angle.