Self-Recurrent Learning and Gap Sample Feature Synthesis-Based Object Detection Method

Table 2

Effect of each component.

SLFF	REAML	GSFF	Recall	Precision	[email protected]
✓			60.32	40.62	55.66	43.20
	✓		73.07	52.70	72.55	59.02
		✓	60.82	41.96	58.92	44.82
✓	✓		71.17	49.97	70.10	56.52
✓	✓	✓	77.95	61.95	79.07	65.00

Results are reported on COCO2017. SLFF: self-recurrent learning-based feature fusion, REAML: residual enhancement architecture-based multichannel, and GSFF: and gap sample-based feature fusion.