Rui Dai

Rui Dai

Auto-generated profile, based on publicly available information

Rui Dai

Inria Université Côte d'Azur 2004 Route des Lucioles 06902 Valbonne

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 November 2021 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
SDSrijan Das
Srijan Das
RDRui Dai
Rui Dai
DYDi Yang
Di Yang
Francois BremondFrancois Bremond
Francois Bremond
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
Verified
Journal Article
Closed access
RDRui Dai
Rui Dai
SDSrijan Das
Srijan Das
SSSaurav Sharma
Saurav Sharma
LMLuca Minciullo
Luca Minciullo
LGLorenzo Garattoni
Lorenzo Garattoni
GFGianpiero Francesca
Gianpiero Francesca
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
Verified
Journal Article
Open access
PDF file
DTDan Tang
Dan Tang
RDRui Dai
Rui Dai
LTLiu Tang
Liu Tang
XLXiong Li
Xiong Li
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.

29 December 2019 in Sensors
Verified
Journal Article
Open access
PDF file
SZSijia Zhan
Sijia Zhan
DTDan Tang
Dan Tang
JMJianping Man
Jianping Man
RDRui Dai
Rui Dai
WXWang Xiyin
Wang Xiyin
Low-rate denial of service (LDoS) attacks reduce the quality of network service by sending periodical packet bursts to the bottleneck routers. It is difficult to detect by counter-DoS mechanisms due to its stealthy and low average attack traffic behavior. In this paper, we propose an anomaly detection method based on adaptive fusion of multiple features (MAF-ADM) for LDoS attacks. This study is based on the fact that the time-frequency joint distribution of the legitimate transmission control protocol (TCP) traffic would be changed under LDoS attacks. Several statistical metrics of the time-frequency joint distribution are chosen to generate isolation trees, which can simultaneously reflect the anomalies in time domain and frequency domain. Then we calculate anomaly score by fusing the results of all isolation trees according to their ability to isolate samples containing LDoS attacks. Finally, the anomaly score is smoothed by weighted moving average algorithm to avoid errors caused by noise in the network. Experimental results of Network Simulator 2 (NS2), testbed, and public datasets (WIDE2018 and LBNL) demonstrate that this method does detect LDoS attacks effectively with lower false negative rate.

1 May 2016
Verified
Conference Paper
Closed access
XWXiong Wang
Xiong Wang
WWWei Wang
Wei Wang
CZChunhui Zeng
Chunhui Zeng
RDRui Dai
Rui Dai
SWSheng Wang
Sheng Wang
SXShizhong Xu
Shizhong Xu
Content-Centric Networking (CCN) is a novel networking paradigm that treats the named contents, not the hosts, as the first-class citizens of the network. In the forwarding plane, CCN employs a stateful forwarding scheme, which maintains per-packet state information in Pending Interest Table (PIT). By employing stateful forwarding, CCN enables native support for content requests aggregation and multicast. However, the stateful forwarding scheme requires large-sized PITs with extremely high access speed to store per-packet state information, leading to scalability issue. To overcome the issue, this paper proposes a Hybrid forwarding scheme based on content POPularity (HyPOP) for CCN. HyPOP classifies the contents into popular and unpopular contents, and uses the stateful and Bloom Filter based stateless forwarding schemes to forward the popular and unpopular contents, respectively. The mathematical analysis results demonstrate that if PITs only store state information for popular contents, the small-sized PITs are sufficient for achieving satisfactory forwarding performance. Furthermore, the extensive simulation results also verify that HyPOP can reduce the size of PIT significantly and achieve promising forwarding performance.

11 July 2022 in arXiv
Verified
Preprint
Open access
PDF file
RDRui Dai
Rui Dai
SDSrijan Das
Srijan Das
Francois BremondFrancois Bremond
Francois Bremond
Action detection is an essential and challenging task, especially for densely labelled datasets of untrimmed videos. There are many real-world challenges in those datasets, such as composite action, co-occurring action, and high temporal variation of instance duration. For handling these challenges, we propose to explore both the class and temporal relations of detected actions. In this work, we introduce an end-to-end network: Class-Temporal Relational Network (CTRN). It contains three key components: (1) The Representation Transform Module filters the class-specific features from the mixed representations to build graph-structured data. (2) The Class-Temporal Module models the class and temporal relations in a sequential manner. (3) G-classifier leverages the privileged knowledge of the snippet-wise co-occurring action pairs to further improve the co-occurring action detection. We evaluate CTRN on three challenging densely labelled datasets and achieve state-of-the-art performance, reflecting the effectiveness and robustness of our method.