Recent technology and market trends have demanded the significant need for feasible solutions to video/camera systems and analytics. This paper provides a comprehensive account on theory and application of intelligent video systems and analytics. It highlights the video system architectures, tasks, and related analytic methods. It clearly demonstrates that the importance of the role that intelligent video systems and analytics play can be found in a variety of domains such as transportation and surveillance. Research directions are outlined with a focus on what is essential to achieve the goals of intelligent video systems and analytics.

This paper proposes the control system for 3-D locomotion of a humanoid biped robot based on a biological approach. The muscular system in the human body and the neural oscillator for generating locomotion signals are adapted in this paper. We extend the neuro-locomotion system for modeling a multiple neuron system, where motoric neurons represent the muscular system and sensoric neurons represent the sensor system inside the human body. The output signals from coupled neurons representing the angle joint level are controlled by gain neurons that represent the energy burst for driving the joint in each motor. The direction and the length of step in robot locomotion can be adjusted by command neurons. In order to form the locomotion pattern, we apply multiobjective evolutionary computation to solve the multiobjective problem when optimizing synapse weights between the motoric neurons. We use recurrent neural network (RNN) for the stabilization system required for supporting locomotion. RNN generates a dynamic weight synapse value between the sensoric neuron and the motoric neuron. The effectiveness of our system is demonstrated in open dynamic engine computer simulation and in a real robot application that has 12 degrees of freedom (DoFs) in legs and four DoFs in hands.

This paper proposes a localization method using multiresolution maps for the navigation of multiple mobile robots based on formation behaviors. The remote control of multiple mobile robots is one the most important tasks in robotics to realize distributed remote monitoring in unknown and/or dynamic environments. However, it is very difficult for a human operator to control multiple mobile robots separately at the same time. Therefore, autonomous formation behaviors of multiple robots are required to reduce mental and physical loads of the human operator. If each mobile robot can estimate the self-position or relative position in a group, it is easier for multiple mobile robots to realize formation behaviors. First, we propose a method of simultaneous localization and mapping based on a grid approach. Next, we explain how to share the build map among multiple mobile robots, and propose a self-localization method based on multiresolution maps. Furthermore, we explain the formation behaviors of multiple mobile robots. Finally, we show several experimental results, and discuss the effectiveness of the proposed method.

Activity recognition which aims to accurately distinguish human actions in complex environments plays a key role in human-robot/computer interaction. However, long-lasting and similar actions will cause poor feature sequence extraction and thus lead to a reduction of the recognition accuracy. We propose a novel discriminative deep model (D3D-LSTM) based on 3D-CNN and LSTM for both single-target and interaction action recognition to improve the spatiotemporal processing performance. Our models have several notable properties: 1) A real-time feature fusion method is used to obtain a more representative feature sequence through composition of local mixtures for enhancing the performance of discriminating similar actions; 2) We introduce an improved attention mechanism that focuses on each frame individually by assigning different weights in real-time; 3) An alternating optimization strategy is proposed for our model to obtain parameters with the best performance. Because the proposed D3D-LSTM model is efficient enough to be used as a detector that recognizes various activities, a Real-set database is collected to evaluate action recognition in complex real-world scenarios. For long-term relations, we update the present memory state via the weight-controlled attention module that enables the memory cell to store better long-term features. The densely connected bimodal modal makes local perceptrons of 3D-Conv motion-aware and stores better short-term features. The proposed D3D-LSTM model has been evaluated through a series of experiments on the Real-set and open-source datasets, i.e. SBU-Kinect and MSR-action-3D. Experimental results show that the proposed D3D-LSTM model achieves new state-of-the-art results, including pushing the average rate of the SBU-Kinect to 92.40% and the average rate of the MSR-action-3D to 95.40%.