MOT Challenge

LINF1_16: Improving Multi-Frame Data Association with Sparse Representations for Robust Near-Online Multi-Object Tracking

MOT16-01

MOT16-01

MOT16-03

MOT16-06

MOT16-07

MOT16-08

MOT16-12

MOT16-14

Benchmark:

MOT16 |

Short name:

LINF1_16

Detector:

Public

Description:

n/a

Reference:

L. Fagot-Bouquet, R. Audigier, Y. Dhome, F. Lerasle. Improving Multi-Frame Data Association with Sparse Representations for Robust Near-Online Multi-Object Tracking. In ECCV, 2016.

Last submitted:

March 11, 2016 (5 years ago)

Published:

March 16, 2016 at 15:51:48 CET

Submissions:

1

Project page / code:

n/a

Open source:

No

Hardware:

2.7 GHZ, 8 Core

Runtime:

4.2 Hz

Benchmark performance:

Sequence	MOTA	IDF1	HOTA	MOTP	MT	ML	FP	FN	Rcll	Prcn	AssA	DetA	AssRe	AssPr	DetRe	DetPr	LocA	FAF	ID Sw.	Frag
MOT16	41.0	45.7	34.7	74.8	88 (11.6)	389 (51.3)	7,896	99,224	45.6	91.3	36.3	33.3	38.1	75.7	35.3	70.8	78.5	1.3	430 (9.4)	963 (21.1)

Detailed performance:

Sequence	MOTA	IDF1	HOTA	MOTP	MT	ML	FP	FN	Rcll	Prcn	AssA	DetA	AssRe	AssPr	DetRe	DetPr	LocA	FAF	ID Sw.	Frag
MOT16-01	29.3	43.6	31.4	71.8	5	12	44	4,464	30.2	97.8	45.2	21.8	47.6	72.6	22.3	72.3	76.7	0.1	12	39
MOT16-03	47.7	48.2	36.6	74.8	29	33	5,999	48,463	53.6	90.3	34.9	38.5	36.3	76.6	41.6	70.0	78.4	4.0	228	513
MOT16-06	40.0	50.3	38.0	73.4	27	130	244	6,637	42.5	95.3	46.3	31.4	51.9	69.3	32.6	72.7	77.5	0.2	42	95
MOT16-07	37.8	45.5	32.5	73.1	6	20	585	9,508	41.7	92.1	34.9	30.4	36.6	73.6	31.8	70.3	77.0	1.2	54	137
MOT16-08	30.8	36.6	30.1	78.8	6	28	370	11,160	33.3	93.8	34.6	26.2	37.0	75.3	27.2	76.5	81.5	0.6	53	73
MOT16-12	41.6	55.9	41.7	76.6	16	54	337	4,493	45.8	91.9	51.4	33.9	53.7	78.0	36.0	72.3	79.7	0.4	11	32
MOT16-14	14.3	28.1	21.5	74.7	0	189	822	14,990	18.9	80.9	29.0	15.9	30.9	70.2	16.4	70.6	78.1	1.1	26	64

Raw data:

LINF1_16