12 November 2021 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Many attempts have been made towards combining RGB and 3D poses for the recognition of Activities of Daily Living (ADL). ADL may look very similar and often necessitate to model fine-grained details to distinguish them. Because the recent 3D ConvNets are too rigid to capture the subtle visual patterns across an action, this research direction is dominated by methods combining RGB and 3D Poses. But the cost of computing 3D poses from RGB stream is high in the absence of appropriate sensors. This limits the usage of aforementioned approaches in real-world applications requiring low latency. Then, how to best take advantage of 3D Poses for recognizing ADL? To this end, we propose an extension of a pose driven attention mechanism: Video-Pose Network (VPN), exploring two distinct directions. One is to transfer the Pose knowledge into RGB through a feature-level distillation and the other towards mimicking pose driven attention through an attention-level distillation. Finally, these two approaches are integrated into a single model, we call VPN++ . It is worth noting that VPN++ exploits the pose embeddings at training via distillation but not at inference. We show that VPN++ is not only effective but also provides a high speed up and high resilience to noisy Poses. VPN++, with or without 3D Poses, outperforms the representative baselines on 4 public datasets. Code is available at https://github.com/srijandas07/vpnplusplus .
25 April 2022 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Designing activity detection systems that can be successfully deployed in daily-living environments requires datasets that pose the challenges typical of real-world scenarios. In this paper, we introduce a new untrimmed daily-living dataset that features several real-world challenges: Toyota Smarthome Untrimmed (TSU). TSU contains a wide variety of activities performed in a spontaneous manner. The dataset contains dense annotations including elementary, composite activities and activities involving interactions with objects. We provide an analysis of the real-world challenges featured by our dataset, highlighting the open issues for detection algorithms. We show that current state-of-the-art methods fail to achieve satisfactory performance on the TSU dataset. Therefore, we propose a new baseline method for activity detection to tackle the novel challenges provided by our dataset. This method leverages one modality (i.e. optic flow) to generate the attention weights to guide another modality (i.e RGB) to better detect the activity boundaries. This is particularly beneficial to detect activities characterized by high temporal variance. We show that the method we propose outperforms state-of-the-art methods on TSU and on another popular challenging dataset, Charades.
6 February 2020 in Human-centric Computing and Information Sciences
Low-rate denial of service (LDoS) attacks send attacking bursts intermittently to the network which can severely degrade the victim system’s Quality of Service (QoS). The low-rate nature of such attacks complicates attack detection. LDoS attacks repeatedly trigger the congestion control mechanism, which can make TCP traffic extremely unstable. This paper investigates the network traffic’ characteristics, in which variance and entropy are used to evaluate the TCP traffic’s characteristics, and the ratio of UDP traffic to TCP traffic (UTR) is also analyzed. Thus, a detection method combining two-step cluster analysis and UTR analysis is proposed. Through two-step cluster analysis which is one of the machine learning algorithms, network traffic is divided into multiple clusters and then clusters subjected to LDoS attacks are determined using UTR analysis. NS2 simulation platform and test-bed network environment aim to evaluate the detection approach’s performance. To better assess the effectiveness of the method, public dataset WIDE is also utilized. Experimental results with a good performance prove that the proposed detection approach can accurately detect LDoS attacks.