Human action recognition in dark videos is a challenging task for computer vision due to the low quality of the videos filmed in the dark. Recent studies focused on applying dark enhancement methods to improve the visibility of the video. However, such video processing results in the loss of critical information in the original (un-enhanced) video. Conversely, traditional two-stream methods are capable of learning information from both original and enhanced videos, but it can lead to a significant increase in the computational cost. To address these challenges, we propose a novel knowledge-distillation-based framework, named Dual-Lightness KnowleDge Distillation (DL-KDD), which simultaneously resolves the aforementioned issues by enabling a student model to obtain both original features and light-enhanced knowledge without additional complexity, thus improving the performance of the model and avoiding extra computational cost. Through comprehensive evaluations, the proposed DL-KDD, with only original video required as input during the inference phase, significantly outperforms state-of-the-art methods on the widely-used dark video datasets. The results highlight the excellence of our proposed knowledge-distillation-based framework for dark video human action recognition.

With the increasing popularity of cloud services, cloud service providers not only need to grasp the needs of consumers, but also should be aware of consumers' perceived value and Quality of Experience (QoE). Accordingly, this research addressed the above issues by building up the IP Centrex context of a means-end chain (MEC)-based framework, which can provide good insight into the customer value in purchasing cloud services. This research adopted soft laddering approach based on the MEC framework and QoE questions. Therefore, two sets of data were collected from 38 respondents who have been involved in the cloud services over the last two years were interviewed. An initial set of practices for the value of cloud services was compiled from the interview data, which was converted into the concepts of "attribute", "consequence" and "value". This interview results were coded, aggregated and presented in a set of hierarchical structured value maps. Another interview data was compiled into a set of QoE factors which are of great help to the cloud service providers in marketing. Both the identified factors and MEC framework can provide valuable reference to further research.