paper
-with-
me
Papers
Browse State-of-the-Art
Datasets
Methods
AI Agents
Trends
Digest
🌙
Multiple Object Tracking
벤치마크
Multiple Object Tracking on BDD100K val
27개 결과 ·
⬇ CSV
·
JSON
mMOTA
34.4
37.17
39.95
42.73
45.5
2020-06
2026-11
QDTrack — 36.6 (2020-06-11)
QDTrack — 36.6 (2020-06-11)
QDTrack — 36.6 (2020-06-11)
ByteTrack — 45.5 (2021-10-13)
ByteTrack — 45.5 (2021-10-13)
ByteTrack — 45.5 (2021-10-13)
Unicorn — 41.2 (2022-07-14)
Unicorn — 41.2 (2022-07-14)
Unicorn — 41.2 (2022-07-14)
TETer — 39.1 (2022-07-26)
TETer — 39.1 (2022-07-26)
TETer — 39.1 (2022-07-26)
Adaptive-searching -windows Tracker — 34.4 (2022-08-22)
Adaptive-searching -windows Tracker — 34.4 (2022-08-22)
Adaptive-searching -windows Tracker — 34.4 (2022-08-22)
QDTrack — 42.1 (2022-10-12)
QDTrack — 42.1 (2022-10-12)
QDTrack — 42.1 (2022-10-12)
MOTRv2 — 43.6 (2022-11-17)
MOTRv2 — 43.6 (2022-11-17)
MOTRv2 — 43.6 (2022-11-17)
UNINEXT-H — 44.2 (2023-03-12)
UNINEXT-H — 44.2 (2023-03-12)
UNINEXT-H — 44.2 (2023-03-12)
ContrasTR — 41.7 (2023-11-14)
ContrasTR — 41.7 (2023-11-14)
ContrasTR — 41.7 (2023-11-14)
QDTrack — 36.6 (2020-06-11)
ByteTrack — 45.5 (2021-10-13)
2020-06-11 — QDTrack: mMOTA 36.6
2021-10-13 — ByteTrack: mMOTA 45.5
Rank
Model
mMOTA
mIDF1
TETA
AssocA
Paper
Code
Year
1
ByteTrack
45.5
54.8
55.7
51.5
ByteTrack: Multi-Object Tracking by Associating Every Detection Box
PaddlePaddle/PaddleDetection
·
mikel-brostrom/boxmot
·
ifzhang/ByteTrack
·
+7
2021
2
UNINEXT-H
44.2
56.7
-
-
Universal Instance Perception as Object Discovery and Retrieval
MasterBin-IIAU/UNINEXT
2023
3
MOTRv2
43.6
56.5
-
-
MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors
megvii-research/MOTRv2
·
DanceTrack/DanceTrack
·
bingfengyan/visam
·
+1
2022
4
QDTrack
42.1
54.3
51.3
52.2
QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Tracking
SysCV/qdtrack
·
ethvis/qd-track
2022
5
ContrasTR
41.7
52.9
-
-
Contrastive Learning for Multi-Object Tracking with Transformers
pfdp0/ContrasTR
2023
6
Unicorn
41.2
54.0
-
–
Towards Grand Unification of Object Tracking
masterbin-iiau/unicorn
2022
7
TETer
39.1
53.3
50.8
52.9
Tracking Every Thing in the Wild
SysCV/tet
2022
8
QDTrack
36.6
50.8
47.8
48.5
Quasi-Dense Similarity Learning for Multiple Object Tracking
SysCV/qdtrack
·
syscv/ovtrack
·
SysCV/tet
2020
9
Adaptive-searching -windows Tracker
34.4
56.5
-
-
Tackling multiple object tracking with complicated motions—Re-designing the integration of motion and appearance
2022
10
ByteTrack
45.5
54.8
55.7
51.5
ByteTrack: Multi-Object Tracking by Associating Every Detection Box
PaddlePaddle/PaddleDetection
·
mikel-brostrom/boxmot
·
ifzhang/ByteTrack
·
+7
2021
11
UNINEXT-H
44.2
56.7
-
-
Universal Instance Perception as Object Discovery and Retrieval
MasterBin-IIAU/UNINEXT
2023
12
MOTRv2
43.6
56.5
-
-
MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors
megvii-research/MOTRv2
·
DanceTrack/DanceTrack
·
bingfengyan/visam
·
+1
2022
13
QDTrack
42.1
54.3
51.3
52.2
QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Tracking
SysCV/qdtrack
·
ethvis/qd-track
2022
14
ContrasTR
41.7
52.9
-
-
Contrastive Learning for Multi-Object Tracking with Transformers
pfdp0/ContrasTR
2023
15
Unicorn
41.2
54.0
-
–
Towards Grand Unification of Object Tracking
masterbin-iiau/unicorn
2022
16
TETer
39.1
53.3
50.8
52.9
Tracking Every Thing in the Wild
SysCV/tet
2022
17
QDTrack
36.6
50.8
47.8
48.5
Quasi-Dense Similarity Learning for Multiple Object Tracking
SysCV/qdtrack
·
syscv/ovtrack
·
SysCV/tet
2020
18
Adaptive-searching -windows Tracker
34.4
56.5
-
-
Tackling multiple object tracking with complicated motions—Re-designing the integration of motion and appearance
2022
19
ByteTrack
45.5
54.8
55.7
51.5
ByteTrack: Multi-Object Tracking by Associating Every Detection Box
PaddlePaddle/PaddleDetection
·
mikel-brostrom/boxmot
·
ifzhang/ByteTrack
·
+7
2021
20
UNINEXT-H
44.2
56.7
-
-
Universal Instance Perception as Object Discovery and Retrieval
MasterBin-IIAU/UNINEXT
2023
1–20 / 27
다음 →
페이지당
10
20
50
100