Search results

Conventional learning-based visual odometry (VO) systems usually use convolutional neural networks (CNN) to extract features, where some important context-related and attention-holding global features might be ignored. Without essential global features, VO system will be sensitive to various environmental perturbations. The purpose of this paper is to design a novel learning-based framework that aims to improve accuracy of learning-based VO without decreasing the generalization ability.

Instead of CNN, a context-gated convolution is adopted to build an end-to-end learning framework, which enables convolutional layers that dynamically capture representative local patterns and composes local features of interest under the guidance of global context. In addition, an attention mechanism module is introduced to further improve learning ability and enhance robustness and generalization ability of the VO system.

The proposed system is evaluated on the public data set KITTI and the self-collected data sets of our college building, where it shows competitive performance compared with some classical and state-of-the-art learning-based methods. Quantitative experimental results on the public data set KITTI show that compared with CNN-based VO methods, the average translational error and rotational error of all the test sequences are reduced by 45.63% and 37.22%, respectively.

The main contribution of this paper is that an end-to-end deep context gate convolutional VO system based on lightweight attention mechanism is proposed, which effectively improves the accuracy compared with other learning-based methods.

Apart from, the smart edge computing (EC) robot (SECR) provides the tools to manage Internet of things (IoT) services in the edge landscape by means of real-world test-bed designed in ECR. Eventually, based on the results from two experiments held in little constrained condition, such as the maximum data size is 2GB, the performance of the proposed techniques demonstrate the effectiveness, scalability and performance efficiency of the proposed IoT model.

Certainly, the proposed SECR is trying primarily to take over other traditional static robots in a centralized or distributed cloud environment. One aspect of representation of the proposed edge computing algorithms is due to challenge to slow down the consumption of time which happened in an artificial intelligence (AI) robot system. Thus, the developed SECR trained by tiny machine learning (TinyML) techniques to develop a decentralized and dynamic software environment.

Specifically, the waste time of SECR has actually slowed down when it is embedded with Edge Computing devices in the demonstration of data transmission within different paths. The TinyML is applied to train with image data sets for generating a framework running in the SECR for the recognition which has also proved with a second complete experiment.

The work presented in this paper is the first research effort, and which is focusing on resource allocation and dynamic path selection for edge computing. The developed platform using a decoupled resource management model that manages the allocation of micro node resources independent of the service provisioning performed at the cloud and manager nodes. Besides, the algorithm of the edge computing management is established with different path and pass large data to cloud and receive it. In this work which considered the SECR framework is able to perform the same function as that supports to the multi-dimensional scaling (MDS).

Micro-video platforms have gained attention in recent years and have also become an important new channel for merchants to advertise their products. Since little research has studied micro-video advertising, this paper aims to fill the research gap by exploring the determinants of micro-video advertising clicks. We form a micro-video advertising click prediction model and demonstrate the effectiveness of the multimodal information extracted from the advertisement producers, commodities being sold and micro-video contents in the prediction task.

A multimodal analysis framework was conducted based on real-world micro-video advertisement datasets. To better capture the relations between different modalities, we adopt a cooperative learning model to predict the advertising clicks.

The experimental results show that the features extracted from different data sources can improve the prediction performance. Furthermore, the combination of different modal features (visual, acoustic, textual and numerical) is also worth studying. Compared to classical baseline models, the proposed cooperative learning model significantly outperforms the prediction results, which demonstrates that the relations between modalities are also important in advertising micro-video generation.

To the best of our knowledge, this is the first study analysing micro-video advertising effects. With the help of our advertising click prediction model, advertisement producers (merchants or their partners) can benefit from generating more effective micro-video advertisements. Furthermore, micro-video platforms can apply our prediction results to optimise their advertisement allocation algorithm and better manage network traffic. This research can be of great help for more effective development of the micro-video advertisement industry.