Online Multiperson Tracking-by-Detection from a Single, Uncalibrated Camera

By Michael D. Breitenstein, Fabian Reichlin, Bastian Leibe, Esther Koller-Meier, Luc Van Gool. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011

668 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/1b7a5f9b9b18ad42608261f7fc7c4879087e7b8a

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

In Global and Planetary Change

Citations in the corpus

Visual SLAM and Structure from Motion in Dynamic Environments (2018, methodology): “Most approaches in foreground initialization make use of the tracking-by-detection scheme [14, 89].”
Deep Network Flow for Multi-object Tracking (2017, neutral): “On-line methods like [8, 11, 15, 39, 41] associate detections of the incoming frame immediately to existing trajectories and are thus appropriate for real-time applications 1 .”
Multiple Hypothesis Tracking Revisited (2015, methodology): “Other works [7, 40] train a classifier for each target and use the classification score for greedy data association or particle filtering.”
DETRAC: A New Benchmark and Protocol for Multi-Object Tracking (2015, background): “, [41, 6, 7, 24, 29, 39, 22, 37, 11], follow the trackingby-detection paradigm, which entails an object detection Figure 1: Precision vs.”
GMMCP tracker: Globally optimal Generalized Maximum Multi Clique problem for multiple object tracking (2015, methodology): “It is worth to mention that tracklets have been previously used as reliable inputs in many tracking algorithm [26, 11, 13, 22].”
People tracking in RGB-D data with on-line boosted target models (2011, methodology): “For this reason, visual tracking systems can achieve good results with methods as simple as independent particle filters with nearest-neighbor data association [11].”
Discrete-continuous optimization for multi-target tracking (2012, methodology): “While early Kalman ﬁltering approaches [21] only model linear target motion, more recent sample-based ﬁlters, such as particle ﬁltering [6, 14], can deal with more complex multi-modal posteriors.”
People detection and tracking from aerial thermal views (2014, background): “1 shows intermediate steps within our tracking pipeline when operating on both frontal and oblique views of humans present in our manually annotated and publicly available dataset.”
Unsupervised Object Discovery and Tracking in Video Collections (2015, background): “Conventional object tracking methods [41] usually require annotations for at least one frame [13, 15, 40], or object detectors trained for target classes in a supervised manner [1, 2, 26].”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, methodology): “Extended Kalman and particle filtering algorithms (Giebel et al. (2004); Breitenstein et al. (2011); Choi et al. (2013)) are widely used models in this context.”
Learning to Track: Online Multi-object Tracking by Decision Making (2015, background): “, Hungarian algorithm [29] in [20, 4] or greedy association [10]).”
Learning People Detectors for Tracking in Crowded Scenes (2013, result): “Additionally, we compare our tracking results with [2] and [6] on the S2.”
In defense of color-based model-free tracking (2015, background): “pedestrian tracking [11, 36, 37, 40]), it is often desirable to build a generic tracker which can readily be used for arbitrary object classes.”
Domain Adaptation for Visual Applications: A Comprehensive Survey (2017, background): “As examples, we have seen the online adaptation for incoming video frames of a generic object detector trained offline on labeled image sets [21, 251].”
Subgraph decomposition for multi-target tracking (2015, neutral): “[6] - - - - - - - - - 2 - 73.”
Robust Online Multi-object Tracking Based on Tracklet Confidence and Online Discriminative Appearance Learning (2014, methodology): “On the other hand, online methods [5, 15, 16, 17, 18] can be applied to real-time applications because they sequentially build trajectories based on frame-by-frame association using online information up to the present frame.”
An online learned CRF model for multi-target tracking (2012, background): “In most previous association based tracking work, appearance models are pre-defined [17, 12] or online learned to discriminate all targets [7, 13] or to discriminate one target with all others [3, 8].”
Online Multi-object Tracking Using CNN-Based Single Object Tracker with Spatial-Temporal Attention Mechanism (2017, methodology): “[5] use targetspecific classifiers to compute the similarity for data association in a particle filtering framework.”