Target segmentation model training method and device and target segmentation method and device

Abstract

The invention relates to a target segmentation model training method and device and a target segmentation method and device. The training method comprises the steps: acquiring the image sample data, wherein each piece of image sample data comprises a target image and a tracking image, the tracking image has a real segmentation label of a target, and the real segmentation label of the target comprises a real value of polar coordinates of a contour of the target; inputting the target image and the tracking image into a target tracking model to obtain a first feature map, wherein the target tracking model is an Ocean model; inputting the first feature map into a target segmentation model to obtain an estimated value of polar coordinates of the contour of the target; calculating a loss function for the target segmentation model based on the estimated value of the polar coordinates of the contour of the target and the true value of the polar coordinates of the contour of the target; and performing joint training on the target tracking model and the target segmentation model based on the loss function for the target tracking model and the loss function for the target segmentation model.

Description

technical field [0001] The present disclosure relates to the field of video technology, and more specifically, to a method and device for training an object tracking and segmentation model, and an object tracking and segmentation method and device. Background technique [0002] Object tracking and segmentation is one of the important technologies in the field of image processing, which is widely used in image / video editing, film and television production and automatic monitoring and other fields. Target tracking technology refers to predicting the size and position of the target object in subsequent frames given the size and position of the target object in the initial frame of a certain video sequence. The target tracking and segmentation technology is based on the target tracking technology, and the pixel-level segmentation results of the target object are given in the prediction of subsequent frames. The traditional target tracking algorithm can only give the position an...

AI Technical Summary

Problems solved by technology

However, due to the limitation of the computing power of the running terminal, the target tracking and segmentation technology based on deep learning is also facing the challenge of speed performance

Images