MOT Challenge

GMPHD_ReId: Occlusion-robust Online Multi-object Visual Tracking using a GM-PHD Filter with a CNN-based Re-identification

MOT16-01

MOT16-01

MOT16-03

MOT16-06

MOT16-07

MOT16-08

MOT16-12

MOT16-14

Benchmark:

MOT16 |

Short name:

GMPHD_ReId

Detector:

Public

Description:

n/a

Reference:

N. Baisa. Occlusion-robust online multi-object visual tracking using a GM-PHD filter with CNN-based re-identification. In Journal of Visual Communication and Image Representation, 2021.

Last submitted:

November 06, 2020 (11 months ago)

Published:

November 30, 2019 at 00:28:16 CET

Submissions:

4

Project page / code:

n/a

Open source:

No

Hardware:

3 GHZ, 1 Core

Runtime:

31.6 Hz

Benchmark performance:

Sequence	MOTA	IDF1	HOTA	MT	ML	FP	FN	Rcll	Prcn	AssA	DetA	AssRe	AssPr	DetRe	DetPr	LocA	FAF	ID Sw.	Frag
MOT16	40.4	50.1	36.0	87 (11.5)	327 (43.1)	6,569	101,251	44.5	92.5	39.5	32.9	43.1	73.7	34.7	72.2	78.7	1.1	789 (17.7)	2,519 (56.6)

Detailed performance:

Sequence	MOTA	IDF1	HOTA	MT	ML	FP	FN	Rcll	Prcn	AssA	DetA	AssRe	AssPr	DetRe	DetPr	LocA	FAF	ID Sw.	Frag
MOT16-01	27.1	43.4	31.3	3	11	182	4,456	30.3	91.4	44.6	22.0	47.1	73.5	22.9	69.0	76.8	0.4	27	84
MOT16-03	47.4	53.6	38.2	23	36	2,407	52,267	50.0	95.6	39.4	37.2	41.8	75.5	38.9	74.4	79.1	1.6	373	1,271
MOT16-06	43.4	48.6	36.0	28	97	687	5,779	49.9	89.3	37.0	35.5	55.5	51.8	38.0	68.1	76.5	0.6	68	256
MOT16-07	36.5	53.5	35.1	5	18	825	9,467	42.0	89.3	40.6	30.6	42.5	74.5	32.4	68.9	77.2	1.7	79	322
MOT16-08	31.6	44.9	34.2	8	24	579	10,781	35.6	91.1	42.2	27.8	44.4	78.2	29.2	74.7	81.3	0.9	90	182
MOT16-12	36.6	55.5	40.9	15	36	913	4,312	48.0	81.4	48.8	34.3	54.7	71.7	38.5	65.2	79.3	1.0	30	84
MOT16-14	17.3	28.7	21.1	5	105	976	14,189	23.2	81.5	25.9	17.3	27.5	71.3	18.3	64.1	76.8	1.3	122	320

Raw data:

GMPHD_ReId