#### DMCA

## Globally-Optimal Greedy Algorithms for Tracking a Variable Number of Objects

### Cached

### Download Links

Citations: | 93 - 1 self |

### Citations

2120 | R.: Fast approximate energy minimization via graph cuts
- Boykov, Veksler, et al.
(Show Context)
Citation Context ...Minimizing C(f) subject to constraints FK is an instance of a minimum cost flow problem [1, 25]. Such problems are similar to max-flow problems (commonly used in vision for solving graph-cut problems =-=[5]-=-), except that edges in the flow network are labeled with a cost as well as capacity. The cost of a flow is defined to be the sum, over all edges, of the cost of each edge multiplied by the flow throu... |

2042 |
flows. Theory, algorithms, and applications
- Ahuja, Magnanti, et al.
- 1993
(Show Context)
Citation Context ...., 0 ≤ fi ≤ 1) This relaxation yields a unit capacity network flow problem whose constraint matrix is totally unimodular, implying that optimal solutions to the relaxed problem will still be integral =-=[1]-=-. In particular, assume that we knew the number of tracks in a video to be K. Let FK denote the set of flow conservation and unit capacity constraints along with the additional constraints FK = { ∈ [0... |

1259 | The Hungarian method for the assignment problem
- Kuhn
- 1955
(Show Context)
Citation Context ...deo data. A popular approach to multi-object tracking is to run a low-level tracker to obtain “tracklets”, and then stitch together tracklets using various graph-based formalisms or greedy heuristics =-=[15, 22, 16, 19, 2]-=-. Such graph-based algorithms include flow-networks [25], linear-programming formulations [14], and matching algorithms [15]. One of the contributions of this paper is to show that with a particular c... |

554 | A discriminatively trained, multiscale, deformable part model
- Felzenszwalb, McAllester, et al.
- 2008
(Show Context)
Citation Context ... we use the same ETHMS video sequence as [25] with 999 frames and ignore detections smaller than 24 pixels as they did. Setup: We ran an “out-of-the-box” pre-trained partbased HOG pedestrian detector =-=[10]-=- with a conservative NMS threshold, generating around 1000 detections per frame of each video. We set the log-likelihood ratio (the local cost ci) of each detection to be the negative score of the lin... |

313 | Bramble: A Bayesian multiple-blob tracker.
- Isard, MacCormick
- 2001
(Show Context)
Citation Context ...ithm, which can boost performance. 73 f 12012. Related Work Classic formulations of multi-object tracking focus on the data association problem of matching instance labels with temporal observations =-=[11, 6, 7, 13]-=-. Many approaches assume manual initialization of tracks and/or a fixed, known number of objects [14]. However, for many real-world tracking problems, such information is not available. A more general... |

193 |
Sonar tracking of multiple targets using joint probabilistic data association,
- Fortmann, Bar-Shalom, et al.
- 1983
(Show Context)
Citation Context ...ithm, which can boost performance. 73 f 12012. Related Work Classic formulations of multi-object tracking focus on the data association problem of matching instance labels with temporal observations =-=[11, 6, 7, 13]-=-. Many approaches assume manual initialization of tracks and/or a fixed, known number of objects [14]. However, for many real-world tracking problems, such information is not available. A more general... |

145 | Global data association for multi-object tracking using network flows.
- Zhang, Li, et al.
- 2008
(Show Context)
Citation Context ...d returns hundreds of correct tracks, as evident by the overlaid track number. Our contribution is grounded in a novel analysis of an integer linear program (ILP) formulation of multi-object tracking =-=[14, 25, 3, 17, 2, 18]-=-. Our work most closely follows the min-cost flow algorithm of [25]. We show that one can exploit the special structure of the tracking problem by using a greedy, successive shortest-path algorithm to... |

140 | Pedestrian detection: A benchmark.
- Dollar, Wojek, et al.
- 2009
(Show Context)
Citation Context ...k death 4 4 32 46 46 4 track birth 46 46 b e track death c d track death Figure 5. We show the results of our algorithm, including estimated track births and deaths, on the Caltech Pedestrian dataset =-=[8]-=-. We show typical results on the ETHMS dataset in Fig.1. ments, this decreased computation time by three orders of magnitude. 6. Experimental Results Datasets: Most benchmarks for multi-object trackin... |

139 | An efficient implementation of a scaling minimum-cost flow algorithm
- Goldberg
- 1997
(Show Context)
Citation Context ...duces state-of-the-art tracks. Our work relies heavily on the min-cost flow network introduced for temporal data association in [25]. We compare our results with the min-cost solver used in that work =-=[12]-=-, and verified that our O(KN log N) algorithm produces identical results, and that our approximate O(KN) algorithm produces near-identical results when properly tuned. In concurrent work, Berclaz et a... |

123 | Multiple object tracking using k-shortest paths optimization,”
- Berclaz, Fleuret, et al.
- 2011
(Show Context)
Citation Context ...results, and that our approximate O(KN) algorithm produces near-identical results when properly tuned. In concurrent work, Berclaz et al. describe a O(KN log N) algorithm for multi-object tracking in =-=[4]-=-. It is similar in many respects with some differences: Our graph representation has a pair of nodes for each detection. This allows us to explicitly model object dynamics through transition costs, an... |

117 |
Depth and appearance for mobile scene analysis,”
- Ess, Leibe, et al.
- 2007
(Show Context)
Citation Context ....g., the number of objects and their track births and deaths) must be estimated. We show the output of an efficient, linear-time algorithms for solving this computational problem on the ETHMS dataset =-=[9]-=-. In this video clip our method returns hundreds of correct tracks, as evident by the overlaid track number. Our contribution is grounded in a novel analysis of an integer linear program (ILP) formula... |

101 | Coupled detection and trajectory estimation for multi-object tracking. In
- Leibe, Schindler, et al.
- 2007
(Show Context)
Citation Context ...deo data. A popular approach to multi-object tracking is to run a low-level tracker to obtain “tracklets”, and then stitch together tracklets using various graph-based formalisms or greedy heuristics =-=[15, 22, 16, 19, 2]-=-. Such graph-based algorithms include flow-networks [25], linear-programming formulations [14], and matching algorithms [15]. One of the contributions of this paper is to show that with a particular c... |

84 | A linear programming approach for multiple object tracking.
- Jiang, Fels, et al.
- 2007
(Show Context)
Citation Context ...d returns hundreds of correct tracks, as evident by the overlaid track number. Our contribution is grounded in a novel analysis of an integer linear program (ILP) formulation of multi-object tracking =-=[14, 25, 3, 17, 2, 18]-=-. Our work most closely follows the min-cost flow algorithm of [25]. We show that one can exploit the special structure of the tracking problem by using a greedy, successive shortest-path algorithm to... |

75 |
Multi-object tracking through simultaneous long occlusions and split-merge conditions.
- Perera, Srinivas, et al.
- 2006
(Show Context)
Citation Context ...deo data. A popular approach to multi-object tracking is to run a low-level tracker to obtain “tracklets”, and then stitch together tracklets using various graph-based formalisms or greedy heuristics =-=[15, 22, 16, 19, 2]-=-. Such graph-based algorithms include flow-networks [25], linear-programming formulations [14], and matching algorithms [15]. One of the contributions of this paper is to show that with a particular c... |

58 | Estimating tracking sources and sinks
- Stauffer
- 2003
(Show Context)
Citation Context |

49 | Robust visual tracking for multiple targets. In
- Cai, Freitas, et al.
- 2006
(Show Context)
Citation Context ...ithm, which can boost performance. 73 f 12012. Related Work Classic formulations of multi-object tracking focus on the data association problem of matching instance labels with temporal observations =-=[11, 6, 7, 13]-=-. Many approaches assume manual initialization of tracks and/or a fixed, known number of objects [14]. However, for many real-world tracking problems, such information is not available. A more general... |

48 | Multi-object tracking through occlusions by local tracklets filtering and global tracklets association with detection responses.
- Xing, Ai
- 2009
(Show Context)
Citation Context ...fect of adding a NMS step within 1207Algorithm Detection rate False positive per frame [9]’s stereo algorithm 47 1.5 [25]’s algorithm 1 68.3 0.85 [25]’s algorithm 2 with occlusion handling 70.4 0.97 =-=[23]-=-’s two-stage algorithm with occlusion handling 75.2 0.939 Our DP 76.6 0.85 Our DP+NMS 79.8 0.85 Table 2. Our algorithm performance compared to the previous state-of-the-art on the ETHMS dataset. Pleas... |

42 | Evaluating multi-object tracking
- Smith, Ba, et al.
- 2005
(Show Context)
Citation Context ...us to compare with a wide body of related work on these datasets. To directly score tracker accuracy, various other criteria (such as track fragmentation, identity switching, etc.) have been proposed =-=[21, 20, 25]-=-. We also use track identity to evaluate our algorithms below. Approximation quality: We have described three different algorithms for solving the minimum cost flow problem. Figure 6 shows the flow co... |

40 | Improving data association by joint modeling of pedestrian trajectories and groupings
- Pellegrini, Ess, et al.
(Show Context)
Citation Context ...d returns hundreds of correct tracks, as evident by the overlaid track number. Our contribution is grounded in a novel analysis of an integer linear program (ILP) formulation of multi-object tracking =-=[14, 25, 3, 17, 2, 18]-=-. Our work most closely follows the min-cost flow algorithm of [25]. We show that one can exploit the special structure of the tracking problem by using a greedy, successive shortest-path algorithm to... |

32 | Multiple object tracking using flow linear programming. In Winter-PETS,
- Berclaz, Fleuret, et al.
- 2009
(Show Context)
Citation Context |

28 |
Globally optimal multi-target tracking on a hexagonal lattice
- Andriyenko, Schindler
(Show Context)
Citation Context |

27 | Multiple target tracking in world coordinate with single, minimally calibrated camera. In: ECCV.
- Choi, Savarese
- 2010
(Show Context)
Citation Context |

19 |
PETS metrics: Online performance evaluation service.
- Young, Ferryman
- 2005
(Show Context)
Citation Context ...cal results on the ETHMS dataset in Fig.1. ments, this decreased computation time by three orders of magnitude. 6. Experimental Results Datasets: Most benchmarks for multi-object tracking (e.g., PETS =-=[24]-=-) are designed for stationary cameras. We are interested in moving camera applications, and so use the Caltech Pedestrian dataset [8] and ETHMS dataset [9] to evaluate our algorithms. The Caltech data... |

9 |
Target tracking with incomplete detection.
- Ma, Yu, et al.
- 2009
(Show Context)
Citation Context |

2 |
Evaluation of algorithms for tracking multiple objects in video
- Perera, Hoogs, et al.
- 2006
(Show Context)
Citation Context ...us to compare with a wide body of related work on these datasets. To directly score tracker accuracy, various other criteria (such as track fragmentation, identity switching, etc.) have been proposed =-=[21, 20, 25]-=-. We also use track identity to evaluate our algorithms below. Approximation quality: We have described three different algorithms for solving the minimum cost flow problem. Figure 6 shows the flow co... |