EURASIP Journal on Advances in Signal Processing

Table 2 Pixel level confusion percentages for our LRC/CRF/OS method on the MSRC-21 dataset. The mean pixel level accuracy is 76.9%.

From: Scene Segmentation with Low-Dimensional Semantic Representations and Conditional Random Fields

										Inferred class
	Building	Grass	Tree	Cow	Sheep	Sky	Aeroplane	Water	Face	Car	Bike	Flower	Sign	Bird	Book	Chair	Road	Cat	Dog	Body	Boat
True class
Building	62.8	1.7	10.5	0.1	0	3.2	0.2	2.1	2.6	0.4	3.5	0	0	0.1	2.3	1.4	7.6	0	0	1.2	0.3
Grass	0.9	94.5	1.4	0.1	0.3	0.1	0.1	0.6	0	0	0	0.3	0	0	0	0	1.6	0	0	0.2	0
Tree	2.2	6.7	82.1	0	0	2.6	0.2	0	0	0	0	4.2	0	0	0	0	0.6	0	0	1.3	0.1
Cow	0	21.4	0.3	70.1	4.0	0	0	0.8	0	0	0	0.1	0	0	0	0	0.5	0.4	0.1	2.2	0
Sheep	0	19.8	0	0.1	79.7	0	0	0	0	0	0.2	0	0	0	0	0	0.2	0	0	0	0
Sky	1.7	0.1	0.5	0	0	92.2	0.1	0	0	0	0	0	0	5.1	0	0	0.3	0	0	0	0
Aeroplane	23.6	10.4	0.8	0.8	0	5.0	56.4	0	0	0	0	0	0	0	0	0	3.0	0	0	0	0
Water	3.3	3.4	6.3	0	0	2.8	0.2	66.5	0	0	0	0	0	2.8	0.9	0	13.4	0	0	0.4	0.1
Face	4.2	0.4	4.0	0	0	1.1	0	0.3	72.4	0	0	0.3	0	0	3.9	0.1	0.1	0	6.9	6.2	0
Car	12.1	0	2.0	0	0	0.1	0	0.0	0	62.6	0	0	0	0.6	0	0.2	15.0	5.8	0	0	1.6
Bike	5.7	0.3	0.8	0	0	0	0	0	0	0.4	70.1	0	0	0	0	9.0	13.6	0	0	0	0 00
Flower	0	0.4	1.9	0	0	0	0	0	0	0	0	97.6	0	0	0	0	0	0	0	0.1	0
Sign	40.4	0.1	1.4	0	0	1.6	0	0	0	0	0	4.2	42.7	0	9.0	0	0.3	0.1	0	0.3	1.0
Bird	4.6	11.5	3.5	0.9	0.8	3.2	3.0	5.9	0	3.5	0	6.5	0	43.5	0	7.8	4.3	0	0	0	0
Book	3.1	0.1	0	0	0	0.2	0	0	0.2	0	0	0	0	0	96.1	0	0.1	0	0	0.2	0
Chair	0.4	19.3	7.3	4.4	0	0	0	3.6	0	0	2.1	0	0.9	1.2	2.4	53.5	4.8	0	0	0	0
Road	4.0	2.0	1.1	0	0	1.7	0.0	8.6	0.1	0.7	1.0	0	0	0	0	0.1	80.2	0	0	0.4	0
Cat	3.0	0.1	0.3	0	0	0.6	0	5.4	0	0	0.4	0	0	4.0	0	0	12.1	74.1	0	0	0
Dog	2.4	5.1	3.0	7.4	0	0.5	0	16.1	7.2	0	4.0	0	0	1.7	0.4	0	10.7	3.8	35.1	2.7	0
Body	7.1	6.5	10.7	0.7	0	0.4	0	6.3	4.1	0	0	1.0	0	0.4	5.7	1.9	3.8	0	4	45.8	1.4
Boat	29.8	0.1	0	0	0	2.2	0.5	28.8	0	1.5	5.7	0	0	1.8	0	0	9.6	0	0	2.2	17.9

Back to article page