MOT Challenge

Video not available.

ETH-Linthescher

Rendering of new sequences is currently deactivated due to heavy load.

AVG-TownCentre

Rendering of new sequences is currently deactivated due to heavy load.

Rendering of new sequences is currently deactivated due to heavy load.

Rendering of new sequences is currently deactivated due to heavy load.

Rendering of new sequences is currently deactivated due to heavy load.

Rendering of new sequences is currently deactivated due to heavy load.

Benchmark:

MOT15 |

Short name:

HAM_INTP15

Detector:

Public

Description:

https://github.com/yyc9268/Deep-TAMA

Reference:

Y. Yoon, A. Boragule, Y. Song, K. Yoon, M. Jeon. Online Multi-Object Tracking with Historical Appearance Matching and Scene Adaptive Detection Filtering. In IEEE AVSS, 2018.

Last submitted:

July 15, 2018 (5 years ago)

Published:

July 15, 2018 at 18:08:48 CET

Submissions:

Project page / code:

https://github.com/yyc9268/Deep-TAMA

Open source:

Yes

Hardware:

3.7 GHZ, 1 Core, TITAN X

Runtime:

18.7 Hz

Benchmark performance:

Sequence	MOTA	IDF1	HOTA	MT	ML	FP	FN	Rcll	Prcn	AssA	DetA	AssRe	AssPr	DetRe	DetPr	LocA	FAF	ID Sw.	Frag
MOT15	28.6	41.4	30.4	72 (10.0)	317 (44.0)	7,485	35,910	41.6	77.3	33.0	28.3	36.2	69.0	31.8	59.1	75.2	1.3	460 (11.1)	1,038 (25.0)

Detailed performance:

Sequence	MOTA	IDF1	HOTA	MT	ML	FP	FN	Rcll	Prcn	AssA	DetA	AssRe	AssPr	DetRe	DetPr	LocA	FAF	ID Sw.	Frag
ADL-Rundle-1	16.7	38.3	28.9	5	13	1,949	5,770	38.0	64.5	35.0	24.2	37.9	72.2	29.4	49.8	76.3	3.9	32	78
ADL-Rundle-3	35.5	39.3	28.1	3	15	442	6,071	40.3	90.3	27.4	29.0	29.5	71.7	30.7	68.7	77.2	0.7	39	56
AVG-TownCentre	20.9	46.5	34.3	30	74	1,762	3,749	47.6	65.9	38.5	30.9	44.2	62.8	37.3	51.7	72.7	3.9	145	301
ETH-Crossing	29.0	46.3	31.0	2	17	28	681	32.1	92.0	40.6	23.7	44.4	73.9	24.7	70.8	78.1	0.1	3	3
ETH-Jelmoli	40.1	60.1	39.4	9	15	269	1,240	51.1	82.8	43.7	35.7	49.2	68.8	39.8	64.4	77.3	0.6	10	33
ETH-Linthescher	19.8	30.6	25.0	9	151	189	6,959	22.1	91.3	37.9	16.6	41.3	73.1	17.1	70.6	77.3	0.2	17	45
KITTI-16	39.5	51.6	31.7	0	3	103	906	46.7	88.5	30.7	32.8	33.3	64.8	35.4	67.1	77.1	0.5	20	42
KITTI-19	20.7	43.6	31.3	4	17	1,311	2,879	46.1	65.3	33.9	29.5	38.2	63.1	35.5	50.2	71.6	1.2	47	193
PETS09-S2L2	46.0	40.9	29.4	5	3	704	4,385	54.5	88.2	23.2	37.7	25.1	64.8	40.4	65.4	73.9	1.6	121	229
TUD-Crossing	63.4	66.5	47.0	4	3	28	370	66.4	96.3	47.8	46.3	50.3	72.6	49.6	71.9	78.2	0.1	5	11
Venice-1	20.6	36.8	28.2	1	6	700	2,900	36.4	70.4	33.1	24.2	34.7	73.6	28.1	54.2	76.2	1.6	21	47

Raw data:

download

ConNextFair

HAM_INTP15

AdTobKF

HAM_INTP15: Historical Appearance Matching and Scene Adaptive Detection Filtering (VisBest)