Enhancing Detection Model for Multiple Hypothesis Tracking
Citations
448 citations
Cites methods from "Enhancing Detection Model for Multi..."
...Other examples of the use of CNNs for feature extraction can be found in [88], where a custom CNN was used to extract appearance features in a Multiple Hypothesis Tracking framework, in [89], whose tracker employed a pretrained region-based CNN [90], or in [91], where a CNN extracted visual features from fish heads, later combined with motion prediction from a Kalman Filter....
[...]
370 citations
239 citations
Cites methods from "Enhancing Detection Model for Multi..."
...on-based MOT Methods Yu et. al [3] proposed the POI algorithm, which conducted a high-performance detector based on Faster R-CNN [4] by adding several extra pedestrian detection datasets. Chen et. al [5] incorporated anenhanceddetection model by simultaneously modeling the detection-scene relation and detection-detection relation, called EDMT. Furthermore, Henschel et. al [6] added a head detection m...
[...]
...d MOTA"IDF1"MOTP" MT" ML# FP# FN# IDS#Hz" Oine MHT-bLSTM [29] 42.1 47.8 75.9 14.9% 44.4% 11637 93172 753 1.8 Quad-CNN [30] 44.1 38.3 76.4 14.6% 44.9% 6388 94775 745 1.8 EDMT [5] 45.3 47.9 75.9 17.0% 39.9% 11122 87890 639 1.8 LMP [31] 48.8 51.3 79.0 18.2% 40.1% 6654 86245 481 0.5 Online CDA-DDAL [32] 43.9 45.1 74.7 10.7% 44.4% 6450 95175 676 - STAM [12] 46.0 50.0 74.9 14.6% 4...
[...]
...ults on MOT17 test dataset. Public Detection Process Method MOTA"IDF1"MOTP" MT" ML# FP# FN# IDS#Hz" Oine MHT-bLSTM [29] 47.5 51.9 77.5 18.2% 41.7% 25981 268042 2069 1.8 EDMT [5] 50.0 51.3 77.3 21.6% 36.3% 32279 247297 2264 1.8 JCC [38] 51.2 54.5 75.9 20.9% 37.0% 25937 247822 1802 - FWT [6] 51.3 47.6 77.0 21.4% 35.2% 24101 247921 2648 - Online DMAN [13] 48.2 55.7 75.9 19.3% 3...
[...]
239 citations
Cites methods from "Enhancing Detection Model for Multi..."
...[56] proposed a multiple hypothesis tracking method by accounting for scene detections and detectiondetection correlations between video frames....
[...]
234 citations
References
10,501 citations
3,564 citations
"Enhancing Detection Model for Multi..." refers methods in this paper
...For training the convolution neural network, we use data from CUHK03[19], Market-1501[32], CUHK01[18], VIPeR[26] and i-LIDS[26]....
[...]
2,417 citations
"Enhancing Detection Model for Multi..." refers methods in this paper
...For training the convolution neural network, we use data from CUHK03[19], Market-1501[32], CUHK01[18], VIPeR[26] and i-LIDS[26]....
[...]
1,812 citations
Additional excerpts
...[20] exploited a novel Figure 2....
[...]
1,499 citations
"Enhancing Detection Model for Multi..." refers background in this paper
...[8] introduced a real-time tracking framework based on adaptive color channels....
[...]