DEEP_TAMA: Online Multiple Pedestrian Tracking with Deep Temporal Appearance Matching Association

MOT17-01-DPM


Benchmark:

MOT17 |

Short name:

DEEP_TAMA

Detector:

Public

Description:

In online multi-target tracking, it is of great importance to model appearance and geometric similarity between pedestrians which have been tracked and appeared in a new frame. The dimension of the inherent feature vector in the appearance model is higher than that in the geometric model, which causes many problems in general. However, the recent success of deep learning-based methods makes it possible to handle high dimensional appearance information successfully. Among many deep networks, the Siamese network with triplet loss is popularly
adopted as an appearance feature extractor. Since the Siamese network can extract features of each input independently, it is possible to update and maintain target-specific features. However, it is not suitable for multi-target settings that require comparison with other inputs. In this paper, to address this issue, we propose a novel track appearance model based on the joint-inference network. The proposed method enables a comparison of two inputs to be used for adaptive appearance modeling, and contributes to disambiguating the process of target-observation matching and consolidating identity consistency. Diverse experimental results support the effectiveness of our method. Our work has been awarded as a 3rd-highest tracker on MOTChallenge19, held in CVPR2019.1 The code is available on https://github.com/yyc9268/Deep-TAMA.

Reference:

Y. Yoon, D. Kim, Y. Song, K. Yoon, M. Jeon. Online Multiple Pedestrians Tracking using Deep Temporal Appearance Matching Association. In Information Sciences, 2020.

Last submitted:

February 07, 2019 (1 year ago)

Published:

February 07, 2019 at 02:56:08 CET

Submissions:

1

Open source:

No

Hardware:

3.7 GHZ, 1 Core(no GPU)

Runtime:

1.5 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
MOT1750.353.576.7453 (19.2)883 (37.5)25,479252,99655.292.41.42,192 (39.7)3,978 (72.1)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
MOT17-01-DPM38.743.971.84111123,82340.795.90.22048
MOT17-01-FRCNN28.340.975.7871,5493,05152.768.73.42645
MOT17-01-SDP44.349.574.6941,2022,35463.577.32.73565
MOT17-03-DPM54.253.575.829292,33145,40756.696.21.6248520
MOT17-03-FRCNN59.657.877.340252,31139,82662.096.61.5186329
MOT17-03-SDP75.970.678.476141,48623,51577.598.21.0183488
MOT17-06-DPM43.255.172.6341097585,85950.388.70.682164
MOT17-06-FRCNN47.354.976.452691,3694,73259.883.71.1115208
MOT17-06-SDP52.258.875.174691,3144,20664.385.21.1108191
MOT17-07-DPM38.543.673.25265389,77742.193.01.179166
MOT17-07-FRCNN31.739.473.73152,0119,37344.578.94.0150290
MOT17-07-SDP46.449.275.514181,0977,85953.589.22.296192
MOT17-08-DPM26.530.480.093947514,97829.192.80.87071
MOT17-08-FRCNN23.130.379.573989215,28027.786.81.46583
MOT17-08-SDP32.636.679.5113451113,62635.593.60.8103136
MOT17-12-DPM41.050.276.519452634,81644.493.60.33748
MOT17-12-FRCNN36.550.077.515456354,83844.285.80.72837
MOT17-12-SDP40.854.478.421428264,27150.784.20.93545
MOT17-14-DPM24.935.874.6610448213,33027.991.40.664107
MOT17-14-FRCNN16.934.371.07743,30811,80336.166.94.4243410
MOT17-14-SDP32.444.972.510652,00910,27244.480.32.7219335

Raw data:


DEEP_TAMA