Detecting loop closures in dynamic environments is a severe challenge for the simultaneous localization and mapping (SLAM) system. Convolutional neural networks (ConvNet) could provide highlevel and abstract representations extracted directly from images as image descriptors. Some novel ConvNet-based methods have been presented. In dynamic environments, they perform better than the stateof-the-art methods which use hand-crafted features. In this paper, (1) We proposed a flexible loop closure detection workflow based on the holistic representations; (2) In this workflow, a post-processing method is applied to the raw holistic ConvNet representations for redundant information compression; (3) In addition, a compression ratio is introduced in (2) to determine how much information will be retained depending on background change and moving objects. We evaluated our workflow in four open datasets. The experimental results demonstrate that the proposed workflow performs better than many state-of-the-art methods and ConvNet-based approaches.

With the development of modern technology and people’s demand for improving living standards, the traditional 2D display cannot meet the needs of people, so the augmented reality (AR) technology which combines virtual objects with real environment arises at the historic moment. AR technology sprouted in the 1960 s, and ushered in rapid development in the 21st century through continuous technical iteration. However, the traditional AR system cannot solve the vergence-accommodation conflict (VAC), to solve which, researchers introduced liquid lens into the optical system of AR. The methods and materials of making liquid lens are different. In this paper, the research progress of liquid lens is summarized from three aspects. Firstly, the research progress of liquid lens is briefly reviewed. Secondly, the classification and typical preparation technology of liquid lens are introduced, and different requirements for performance are described. Finally, the performance and testing method of liquid lens is introduced.