Comparison of metrics

Metric/Variant	RMSE					PLCC					SROCC
Metric/Variant	FovDots	UPIQ	DeepFovea	LIVE-FBT-FCVR	Average⁽¹⁾	FovDots	UPIQ	DeepFovea	LIVE-FBT-FCVR	Average⁽²⁾	FovDots	UPIQ	DeepFovea	LIVE-FBT-FCVR	Average⁽³⁾
Both temporal channels →	0.8938	0.7696	0.9376	0.6481	0.8123	0.7873	0.8599	0.8322	0.728	0.8554	0.8148	0.83	0.8413	0.7254	0.8287
Only sustained channel →	1.143	0.7713	0.9561	0.616	0.8717	0.7298	0.8589	0.8449	0.7534	0.8526	0.7355	0.8272	0.8477	0.7534	0.8246

⁽¹⁾Average RMSE is computed as an average of RMSEs of individual datasets so that each dataset has the same influence on the average RMSE regardless of the number of conditions it contains.

^(2-3)Average correlation coeffcients are computed for the consolidated dataset consisting of all four individual datasets. It means that the average correlation coefficients are dominated by the performance of UPIQ, which contains over 4000 conditions. For that reason, RMSE is more indicative of metric performance.

Both temporal channels

RMSE = 0.8123 PLCC = 0.8554 SROCC = 0.8287

JOD regression: Q = 10 -0.0129523 * M^0.596953

Only sustained channel

RMSE = 0.8717 PLCC = 0.8526 SROCC = 0.8246

JOD regression: Q = 10 -0.0341355 * M^0.510226