Detection methods based on deep convolutional networks search for interest points by constructing response maps using supervised, self-supervised, and unsupervised methods. Supervised methods use ...

Traditional feature matching methods rely on accurate descriptors to describe the attributes of feature points, which may lead to the problem of misalignment when the descriptors are incorrect.

Visual Simultaneous Localization and Mapping (VSLAM) is a crucial technology for autonomous mobile vision robots. However, existing methods often suffer from low localization accuracy and poor ...