Multi-camera activity correlation analysis

By Chen Change Loy, Tao A. Xiang, Shaogang Gong. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009

346 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/3b5b6d19d4733ab606c39c69a889f9e67967f151

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

DeepReID: Deep Filter Pairing Neural Network for Person Re-identification (2014, background): “CUHK03 VIPeR [12] i-LIDS [50] CAVIAR [3] Re-ID 2011 [16] GRID [27] CUHK01 [25] CUHK02 [24] No. of images 13,164 1,264 476 1,220 2,450 500 1,942 7,264 No. of persons 1,360 632 119 72 245 250 971 1,816 (2) Some existing datasets assume a single pair of camera views and their cross-view transforms are ”
Joint Detection and Identification Feature Learning for Person Search (2016, background): “It has many video surveillance applications, such as finding criminals [32], cross-camera person tracking [37], and person activity analysis [23].”
Person Re-Identification by Deep Joint Learning of Multi-Loss Classification (2017, methodology): “We used the benchmarking 10 people splits [ Loy et al. , 2009 ] and the averaged performance.”
Structured Time Series Analysis for Human Action Segmentation and Recognition (2014, background): “However, setting T is a difficult task and how to improve this process is described in Section 3.4.”
Person Re-Identification by Camera Correlation Aware Feature Augmentation (2017, methodology): “We extensively evaluated the proposed approach on five person re-id benchmarks: VIPeR [30], CUHK01 [31], CUHK03 [26], QMUL GRID [32], and Market-1501 [33].”
Omni-Scale Feature Learning for Person Re-Identification (2019, neutral): “Speciﬁcally, different streams have different receptive ﬁelds but are built with the same Lite 3 × 3 CUHK03 [24], DukeMTMC-reID (Duke) [33, 70], MSMT17 [53], VIPeR [10] and GRID [29].”
Person re-identification by Local Maximal Occurrence representation and metric learning (2014, background): “Example pairs of images from the GRID database [31].”
End-to-End Deep Learning for Person Search (2016, methodology): “1http://www.ee.cuhk.edu.hk/˜xgwang/PS/dataset. html Although numerous re-id datasets and algorithms have been proposed in recent years and the performance on these benchmarks have been improved substantially, there is still a big gap with practical applications.”
Person Re-identification: Past, Present and Future (2016, methodology): “A number of datasets for image-based re-ID have been released, and some commonly used datasets are summarized in Table 3 [16], [36], [44], [82], [83], [84], [85], [86], [87], [88], [89], [90] for image-based re-ID. and testing sets.”
Diversity Regularized Spatiotemporal Attention for Video-Based Person Re-identification (2018, background): “This task has drawn increasing attention in recent years due to its importance in applications, such as surveillance [42], activity analysis [32] and tracking [47].”
Efficient PSD Constrained Asymmetric Metric Learning for Person Re-Identification (2015, methodology): “VIPeR [5] and QMUL GRID [17]), resulting in very limited number of positive sample pairs.”
Learning Deep Neural Networks for Vehicle Re-ID with Visual-spatio-Temporal Path Proposals (2017, methodology): “[29] proposed an approach for multi-camera activity correlation analysis which estimates the spatial and temporal topology of the camera network.”
Scalable Person Re-identification on Supervised Smoothed Manifold (2017, methodology): “The proposed Supervised Smoothed Manifold (SSM) is evaluated on five popular benchmarks, including GRID [33, 32], VIPeR [15], PRID450S [41], CUHK03 [24] and Market-1501 [63].”
Person Re-identification by Attributes (2012, background): “Various other complementary aspects of the problem have also been pursued to improve performance, such as improving robustness by combining multiple frames worth of features along a tracklet [1] and learning the topology of the camera network [14] or activity correllations [15] to cut down on the ma”
Pose-Normalized Image Generation for Person Re-identification (2017, methodology): “The other one is the Transfer Learning (TL) setting on VIPeR, CUHK01, GRID, CUHK03, and DukeMTMC-reID dataset.”
Intelligent multi-camera video surveillance: A review (2013, methodology): “The multi-camera activity correlation analysis proposed in (Loy et al., 2009) can improve object re-identification across camera views by providing the contextual information of the temporal and causal relationships between regional activities.”