Tracking Social Groups Within and Across Cameras

@article{Solera2017TrackingSG,
  title={Tracking Social Groups Within and Across Cameras},
  author={Francesco Solera and Simone Calderara and Ergys Ristani and Carlo Tomasi and Rita Cucchiara},
  journal={IEEE Transactions on Circuits and Systems for Video Technology},
  year={2017},
  volume={27},
  pages={441-453}
}
We propose a method for tracking groups from single and multiple cameras with disjointed fields of view. [] Key Method Multicamera group tracking is handled inside the framework by adopting an orthogonal feature encoding that allows the classifier to learn inter- and intra-camera feature weights differently. Experiments were carried out on a novel annotated group tracking data set, the DukeMTMC-Groups data set. Since this is the first data set on the problem, it comes with the proposal of a suitable…

Multi-target Tracking in Multiple Non-overlapping Cameras Using Fast-Constrained Dominant Sets

This paper proposes fast-constrained dominant set clustering (FCDSC), a novel method which is several orders of magnitude faster (close to real time) than existing methods for solving multi-target tracking in non-overlapping cameras and shows that it can be used in conjunction with state-of-the-art video-based re-identification algorithms, to increase their already good performances.

Enhancing the association in multi‐object tracking via neighbor graph

This work first utilizes the spatio-temporal relations produced by the tracking self to efficiently select suitable neighbors for the targets, then employs the graph convolution networks (GCN) to learn the graph features.

Multi-camera Multi-Object Tracking

This work model's the tracking problem as a global graph, and adopts Generalized Maximum Multi Clique optimization problem as the core algorithm to take both across frame and across camera data correlation into account all together.

Deep Neural Network with Extracted Features for Social Group Detection

In the latest years, the use of deep neural networks (DNNs) for learning and detecting has been increased, and in this work, DNNs are used for detecting social groups with extracted features.

Data-driven Online Group Detection Based on Structured Prediction

This paper extends the features of pairwise trajectories with velocity and orientation to obtain more valid information and adopts the adaptive-sampling BCFW algorithm to learn the mapping from trajectories to groups.

Collective Interaction Filtering Approach for Detection of Group in Diverse Crowded Scenes

This work proposes a group detection approach called Collective Interaction Filtering to discover people motion interaction from trajectories that is able to deduce people interaction with the Expectation-Maximization algorithm.

Extensions of dominant sets and their applications in computer vision

This thesis proposes a method for simultaneous clustering and outliers detection, and shows that by properly controlling a regularization parameter which determines the structure and the scale of the underlying problem, the author is in a position to extract groups of dominant sets clusters which are constrained to contain user-specified elements.

Wide-Baseline Multi-Camera Calibration using Person Re-Identification

This work proposes a method that uses ideas from person re-identification (re-ID) for wide-baseline camera calibration, which first employs a re-ID method to associate human bounding boxes across cameras, then converts bounding box correspondences to point correspondences, and finally solves for camera pose using multi-view geometry and bundle adjustment.

Group Split and Merge Prediction With 3D Convolutional Networks

This work formulates this as a video prediction problem, where group splits or merges are predicted given a history of geometric social group shape transformations, and develops a modified C3D model that performs significantly better at predicting the occurrence of splits and merges.

Towards a Principled Integration of Multi-camera Re-identification and Tracking Through Optimal Bayes Filters

This work presents a theoretically principled way of integrating ReID with tracking formulated as an optimal Bayes filter, which conveniently side-steps the need for dataassociation and opens up a direct path from full images to the core of the tracker.

References

SHOWING 1-10 OF 50 REFERENCES

Improving multi-target tracking via social grouping

  • Zhen QinC. Shelton
  • Computer Science
    2012 IEEE Conference on Computer Vision and Pattern Recognition
  • 2012
A general optimization framework that adds social grouping behavior (SGB) to any basic affinity model based upon a simple affinity model is built and shows very promising performance on two publicly available real-world datasets with different tracklet extraction methods.

Target Identity-aware Network Flow for online multiple target tracking

It is shown that automatically detecting and tracking targets in a single framework can help resolve the ambiguities due to frequent occlusion and heavy articulation of targets.

Detection and tracking of groups in crowd

The method embeds plausible human behaviors to predict interactions in a crowd by iteratively minimizing the error between predictions and measurements and outperforms existing group localization techniques on three publicly available datasets.

Probabilistic group-level motion analysis and scenario recognition

This work builds on a mature multi-camera multi-target person tracking system that operates in real-time and derives probabilistic models to analyze individual track motion as well as group interactions to recognize group interactions without making hard decisions about the underlying group structure.

Joint Individual-Group Modeling for Tracking

A novel probabilistic framework that jointly models individuals and groups for tracking, promoting the idea that groups are better modeled if individuals are considered and vice versa and provides convincing results on two recent challenging tracking benchmarks.

Associating Groups of People

A novel people group representation and a group matching algorithm are proposed that addresses changes in the relative positions of people in a group and the latter deals with variations in illumination and viewpoint across camera views.

Learning to Divide and Conquer for Online Multi-target Tracking

This paper claims that the ambiguities in tracking could be solved by a selective use of the features, by working with more reliable features if possible and exploiting a deeper representation of the target only if necessary.

Improving Data Association by Joint Modeling of Pedestrian Trajectories and Groupings

A third-order graphical model is proposed that is able to jointly estimate correct trajectories and group memberships over a short time window and a set of experiments underline the importance of joint reasoning for data association in crowded scenarios.

Structured learning for detection of social groups in crowd

A new Structural SVM-based learning framework able to solve the group detection task by exploiting annotated video data to deduce a sociologically motivated distance measure founded on Hall's proxemics and Granger's causality is built.

Decentralized particle filter for joint individual-group tracking

The proposed strategy factorizes the joint individual-group state space in two dependent subspaces where individuals and groups share the knowledge of the jointindividual-group distribution, promoting the idea that groups are better tracked if individuals are considered, and viceversa.