Rgb D Salient Object Detection On Nju2K

Métriques

Average MAE

S-Measure

max E-Measure

max F-Measure

Résultats

Résultats de performance de divers modèles sur ce benchmark

Tableau comparatif

Nom du modèle	Average MAE	S-Measure	max E-Measure	max F-Measure
progressively-complementarity-aware-fusion	0.059	87.7	92.4	87.2
learning-selective-self-mutual-attention-for	0.053	89.4	92.7	88.9
cola-conditional-dropout-and-language-driven	0.029	93.4	94.7	91.3
cascade-graph-neural-networks-for-rgb-d	0.035	91.1	-	-
rethinking-rgb-d-salient-object-detection	0.046	90.0	93.9	90.0
cross-modal-weighting-network-for-rgb-d	0.046	90.3	-	-
uncertainty-inspired-rgb-d-saliency-detection	0.039	90.2	-	-
siamese-network-for-rgb-d-salient-object	0.040	91.1	94.8	91.3
uc-net-uncertainty-inspired-rgb-d-saliency	0.043	89.7	-	-
rgb-d-salient-object-detection-with-cross	0.044	90.4	-	-
bilateral-attention-network-for-rgb-d-salient	0.039	91.5	94.8	92.0
spsn-superpixel-prototype-sampling-network	0.032	91.8	95.0	92.0
local-background-enclosure-for-rgb-d-salient	0.153	69.5	80.3	74.8
hierarchical-dynamic-filtering-network-for	0.037	91.1	-	-
contrast-prior-and-fluid-pyramid-integration	0.053	87.8	92.6	87.7
bts-net-bi-directional-transfer-and-selection	0.036	92.1	95.4	92.4
rgbd-salient-object-detection-via-deep-fusion	0.205	51.4	72.4	63.2
select-supplement-and-focus-for-rgb-d	0.043	89.9	-	-
uncertainty-inspired-rgb-d-saliency-detection	0.039	90.0	-	-
is-depth-really-necessary-for-salient-object	0.042	90.2	-	91.1
accurate-rgb-d-salient-object-detection-via	0.047	89.4	-	-
depth-quality-aware-salient-object-detection	0.052	89.7	-	-
depth-induced-multi-scale-recurrent-attention	0.051	88.6	92.7	88.6
a2dele-adaptive-and-attentive-depth-distiller	0.051	-	-	-
dformer-rethinking-rgbd-representation	0.023	93.7	96.4	94.6
a-single-stream-network-for-robust-and-real	0.046	89.7	-	90.5
jl-dcf-joint-learning-and-densely-cooperative	0.043	90.3	94.4	90.3