CARF-Net: CNN attention and RNN fusion network for video-based person reidentification

Kansal, Kajal; Subramanyam, A V; Prasad, Dilip K.; Kankanhalli, Mohan

doi:10.1117/1.jei.28.2.023036

Cited by 3 publications

(2 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rank1 is 80.6%, which is 4.4% higher than the method proposed in literature [8]. Rank5 and rank20 algorithms have some improvements compared with other algorithms [13][14][15][16][17].…”

Section: Resultsmentioning

confidence: 83%

Video person reidentification based on neural ordinary differential equations and graph convolution network

Zhang¹,

Huang

Duan

2021

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

Person reidentification rate has become a challenging research topic in the field of computer vision due to the fact that person appearance is easily affected by lighting, posture and perspective. In order to make full use of the continuity of video data on the time line and the unstructured relationship of features, a video person reidentification algorithm combining the neural ordinary differential equation with the graph convolution network is proposed in this paper. First, a continuous time model is constructed by using the ordinary differential equation (ODE) network so as to capture hidden information between video frames. By simulating the hidden space of the hidden variables with the hidden time series model, the hidden information between frames that may be ignored in the discrete model can be obtained. Then, the features of the generated video frames are given to the graph convolution network to reconstruct them. Finally, weak supervision is used to classify the features. Experiments on PRID2011 datasets show that the proposed algorithm can significantly improve person reidentification performance.

show abstract

“…Rank1 is 80.6%, which is 4.4% higher than the method proposed in literature [8]. Rank5 and rank20 algorithms have some improvements compared with other algorithms [13][14][15][16][17].…”

Section: Resultsmentioning

confidence: 83%

Video person reidentification based on neural ordinary differential equations and graph convolution network

Zhang¹,

Huang

Duan

2021

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

show abstract

“…Since the historical trajectory information is typical time series data, the location-prediction algorithm is also a typical time series prediction process. Recently, recurrent neural networks (RNNs) [17] have been adopted in machine translation [18], target recognition [19], video behavior recognition [20], sentiment classification, and image caption generation [21] and show promising performance in processing time series prediction compared with traditional methods. Therefore, RNNs can be used to predict next important locations [22,23].…”

Section: Introductionmentioning

confidence: 99%

Location Extraction and Prediction Method Based on Floating Car Spatial-Temporal Trajectory

Pan

Chong

2020

IJGI

View full text Add to dashboard Cite

Predicting the next important location by mining the user’s historical spatial-temporal trajectory can be done for behavioral analysis and path planning. Since extracting the important location precisely is the premise of next location prediction, an enhanced location extraction algorithm is proposed to meet the requirements of dynamic trajectory via dynamic parameter estimation. To realize the estimation of parameters dynamically, the differences of floating car velocity in terms of spatial distribution and behavior in time distribution are considered in the location extraction algorithm. Then, an improved recurrent neural network (RNN) model is designed to mine the variation law of floating car trajectories to improve the accuracy of important location prediction under different conditions. Different from the traditional prediction model considering only the constraint of distance, the attention mechanism and semantic information are considered simultaneously by the proposed prediction model. Finally, the floating car trajectory of Beijing is selected for our experiments, and the results show that the proposed location extraction algorithm can meet the requirements of a dynamic environment and our model achieves high prediction accuracy.

show abstract

A Soft Sensor Model for Cement Specific Surface Area Based on TCN-ASRU Neural Network

Sun¹,

Zhang²,

Zhao³

et al. 2023

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

CARF-Net: CNN attention and RNN fusion network for video-based person reidentification

Cited by 3 publications

References 60 publications

Video person reidentification based on neural ordinary differential equations and graph convolution network

Video person reidentification based on neural ordinary differential equations and graph convolution network

Location Extraction and Prediction Method Based on Floating Car Spatial-Temporal Trajectory

A Soft Sensor Model for Cement Specific Surface Area Based on TCN-ASRU Neural Network

Contact Info

Product

Resources

About