Recurrent Convolutional Network for Video-Based Person Re-identification

By Niall McLaughlin, Jesús Martínez del Rincón, Paul C. Miller. Computer Vision and Pattern Recognition, 2016

589 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/57052d3357cfedf389e3c7b03d9e5f2772d01038

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Diversity Regularized Spatiotemporal Attention for Video-Based Person Re-identification (2018, background): “Previous video-based person re-identification methods [35, 34, 52] do not model long-range temporal structure because the input video sequences are relatively short.”
Analyzing Neuroimaging Data Through Recurrent Deep Learning Models (2018, unknown): “This combination of convolutional and recurrent DL elements is inspired by previous research, showing that it is generally well-suited to learn the spatial dependency structure of long sequences of input data (Donahue et al., 2015; McLaughlin et al., 2016; Marban et al., 2019).”
Person Re-identification: Past, Present and Future (2016, methodology): “While the two works [114], [120] employ the siamese network for loss computation, Yan et al. [115] and Zheng et al. [21] use the identiﬁcation model which classiﬁes each input video into their respective identities.”
Dynamic Label Graph Matching for Unsupervised Video Re-identification (2017, background): “In recent years, impressive progress has been reported in video based re-ID [34, 20, 37], because video sequences provide rich visual and temporal information and can be trivially obtained by tracking algorithms [11, 12] in practical video surveillance applications.”
Interpretable LSTMs For Whole-Brain Neuroimaging Analyses (2018, background): “that it is generally well-suited to learn the spatio-temporal dependency structure of long sequences of input data [28, 40, 41].”
Discriminability Distillation in Group Representation Learning (2020, background): “It is also beneficial to perform person re-identification [61,39,18,16,17,33,15] from videos.”
A Siamese Long Short-Term Memory Architecture for Human Re-identification (2016, methodology): “Even though a recent work [38] uses RNN for human re-identification, they use it to learn the interaction between multiple frames in a video and not for learning the spatial relationships.”
Dual Attention Matching Network for Context-Aware Feature Sequence Based Person Re-identification (2018, methodology): “Besides, we follow the common practices to augment image dataset by using random horizontal ﬂips and random crops [28], and to augment video dataset by randomly selecting video sub-sequences of 16 consecutive frames.”
Human Semantic Parsing for Person Re-identification (2018, methodology): “[45], adopt simple multi-class classification loss while [49, 29] use a combination of both classification and verification losses.”
End-to-End Comparative Attention Networks for Person Re-Identification (2016, methodology): “Different from the video-based person re-identification methods, such as [20], [39], we investigate the problem of image-based person re-identification in this paper.”
Exploit the Unknown Gradually: One-Shot Video-Based Person Re-identification by Stepwise Learning (2018, background): “Recently, a number of deep learning methods are developed [24, 29, 32, 35, 39, 40].”
Deep Learning for Person Re-Identification: A Survey and Outlook (2020, methodology): “To accurately and automatically capture the temporal information, a recurrent neural network architecture is designed for video-based person Re-ID [129].”
Robust Anchor Embedding for Unsupervised Video Person re-IDentification in the Wild (2018, methodology): “In addition, some CNN-RNN related network structures are also designed for video re-ID task [26, 30, 44].”
Jointly Attentive Spatial-Temporal Pooling Networks for Video-Based Person Re-identification (2017, methodology): “Preprocessing steps included the following actions [20]: Images were converted to YUV color space firstly, and each color channel was normalized to have zero mean and unit variance; Optical flow, both vertical and horizontal, were extracted between each pair of adjoining images using the Lucas-Kanad”
Towards a Visual Privacy Advisor: Understanding and Predicting Privacy Risks in Images (2017, unknown): “Apart from detecting attributes, some works introduce new privacy challenges in vision such as adversarial perturbations [31, 35], privacy-preserving video capture [1, 36, 33, 37], person re-identification [2, 30], avoiding face detection [46, 16], full body re-identification [34] and privacy-sensit”
Multi-scale Deep Learning Architectures for Person Re-identification (2017, methodology): “In contrast, most of the state-of-the-art video-based re-id methods [40, 37, 52, 23, 30, 22] utilized the RNN models such as LSTM to perform temporal/sequence video feature fusion from each frame.”
A Survey of Multi-View Representation Learning (2016, unknown): “[40] combine the visual features from all time-stamps using temporal max pooling to obtain an comprehensive appearance feature for the complete sequence.”
Spindle Net: Person Re-identification with Human Body Region Guided Feature Decomposition and Fusion (2017, background): “In recent years, the video based ReID problem is proposed and much work has been conducted on this field [4, 8, 18, 27].”
Adaptive Transfer Network for Cross-Domain Person Re-Identification (2019, background): “[21] presented a recurrent neural network architecture for video-based person re-identification, which utilized optical flow, recurrent layers and mean-pooling layer to learn video features containing appearance and motion information.”