Convolutional Neural Network (CNN)-based Intrusion Detection System (IDS) incorporated with Federated Learning (FL) facilitates collaborative model training to secure user privacy and enhance data diversity, emerging as a promising solution to defend malicious flows. Yet, FL is vulnerable to trigger backdoor attacks. Existing methods have limited ability on IDS with FL for detecting or repairing a converged attacked model as they either require access to the training process or are dedicated to specific applications. To tackle the limitations, we propose a novel Features Distraction Defense Framework (FDDF) that does not interfere with the model training and requires only a few additional model inferences to protect users from the aforesaid attack. The experiment results show that FDDF can recover the accuracy effectively and preserve the accuracy of normal models.

We detail the results of two experiments that empirically evaluate the efficiency of spatial selection performance of popular pointing and controller based direct 3D interaction within the participants’ maximum arms reach or personal space in an immersive virtual environment. In the first experiment, targets were presented in a circular configuration of 13 targets in accordance with the ISO 9241-9 Fitts’ Law standard, with a target distance of 5, 9 or 18 cm and target width of 0.65, 0.85, or 1.05cm, and with each circular configuration presented in a random distance of 60%, 75% or 90% of maximum arms length of the participant. The targets were centered at eye level, and in a within-subjects protocol participants selected the targets using a pointer at the tip of the controller or at the tip of their index finger on the dominant hand in the pointing based condition. The second experiment presented each target in a similar width and distance configuration as the previous experiment, but each target also varied in depth from one trial to the next between 60%, 75% or 90% of maximum arms reach. The selections were performed in an HTC Vive head mounted display VR world, using the popular Leap Motion gesture tracking device or the Vive controller. The results of both our experiment taken together revealed that with commodity 3D input devices, participants were more efficient in spatial selection performance in the popular controller based condition, as compared to the pointing based condition. Implications for VR developers and consumers are discussed in the latter sections.

Design discussion is a very important course in architecture education. In this paper, we developed a VR based architecture design discussion system that allows members to visualize and discuss the architectural models and to modify the models during discussion. Since the system is designed to work on top of Rhinoceros and Grasshopper, the object database is updated right after the object modification. Members communicate via voice, object manipulations, and mid-air sketching as well as on-surface sketching in the virtual environment. Several tools have been designed to enhance the sense of presence and to make the discussion more effective. We also developed a rollback mechanism to help users intuitively and quickly revert to a previous state of discussion to make some changes or to start a new direction of discussion. To evaluate the system, we conducted an initial user study with 14 participants to assess the user experience, user impression and effectiveness of the system. The feedback from participants were positive and suggested that the system could be effective and useful for supporting architecture design discussion.

Conventional object assembly instructions are usually written or illustrated in a paper manual. Users have to associate these static instructions with real objects in 3D space. In this paper, a novel augmented reality system is presented for a user to interact with objects and instructions. While most related methods pasted obvious markers onto objects for tracking and constrained their orientations or shapes, we adopt a markerless strategy for more intuitive interaction. Based on live information from an off-the-shelf RGB-D camera, the proposed tracking procedure identifies components in a scene, tracks their 3D positions and orientations, and evaluates whether there are combinations of components. According to the detected events and poses, our indication procedure then dynamically displays indication lines, circular arrows and other hints to guide a user to manipulate the components into correct poses. The experiment shows that the proposed system can robustly track the components and respond intuitive instructions at an interactive rate. Most of users in evaluation are interested and willing to use this novel technique for object assembly.

A number of ZigBee based monitoring systems are built on the basis of IOT technology in this work on the perception layer, using temperature/humidity sensors, light sensors and 3-axis accelerometer modules. Wirelessly transmitted to a monitoring center, all the sensed data are collected by a human computer interface. On the application layer in an IOT, simulation experiments are conducted, namely, applications of light sensors to an automated basketball court lighting system, 3-axis accelerometer modules to the monitoring of infant's sleeping posture and accidental fall of the elderly, and temperature/humidity sensors to thermal comfort testing. This research work is validated as an effective way to achieve the aim of power consumption reduction, improve the health care quality and provide a higher comfort level.

Estimating upper body poses from a sequence of depth images is a challenging problem. Lately, the state-of-art work adopted a randomized forest method to label human parts in real time. However, it requires enormous training data to obtain favorable results. In this paper, we propose using a novel two-stage method to estimate the probability maps of upper body parts of users. These maps are then used to evaluate the region fitness of body parts for pose recovery. Experiments show that the proposed method can obtain satisfactory outcome in real time and it requires a moderate size of training data.

VM consolidation has been shown as a promising technique for saving energy costs of a data center. It relies on VM migration to move user applications or jobs onto fewer numbers of physical servers during off peak hour. However, VM migration is a costly operation that could cause several concerns, such as performance degradation and system instability. Most existing works were proposed to minimize the migration cost for dynamic consolidation which migrates VM at the runtime when SLA violation or resource under-utilization is detected. In contrast, this paper aims to proactively prevent VM migration for semi-static VM consolidation by proposing a deadline driven VM placement strategy based on the awareness of the server turn-off time and job execution time. We evaluate our approach using a real HPC cluster trace as well as a set of synthetic generated workloads. The results show our approach can significantly reduce the number of migrations by 70% on the real trace. We also demonstrate that our approach can be resilient to different workload patterns by achieving consistent improvement around 50% over all the synthetic workloads.

In this paper, we present an efficient framework to deform polygonal models for skeleton-driven animation. Standard solutions of skeleton-driven animation, such as linear blend skinning, require intensive artist intervention and focus on primary deformations. The proposed approach can generate both low- and high-frequency surface motions such as muscle deformation and vibrations with little user intervention. Given a surface mesh, we construct a lattice of cubic cells embracing the mesh and we apply lattice-based smooth skinning to drive the surface primary deformation with volume preservation. Lattice shape matching with dynamic particles, in the meantime, is utilized for secondary deformations. Due to the highly parallel lattice structure, the proposed method is liable to GPU computation. Our results show that it is adequate to vividly real-time animation.

In this paper, we present a representation method for motion capture data by exploiting the nearly repeated characteristics and spatiotemporal coherence in human motion. We extract similar motion clips of variable lengths or speeds across the database. Since the coding costs between these matched clips are small, we propose the repeated motion analysis to extract the referred and repeated clip pairs with maximum compression gains. For further utilization of motion coherence, we approximate the subspace-projected clip motions or residuals by interpolated functions with range-aware adaptive quantization. Our experiments demonstrate that the proposed feature-aware method is of high computational efficiency. Furthermore, it also provides substantial compression gains with comparable reconstruction and perceptual errors.