How to use this page?
- Please be patient until all images have loaded completely.
- Move the mouse over the numbers in the table to see the corresponding images.
- Drag the red rectangle in the leftmost image to change the location of the zoom.
- Press and hold any key to temporarily deactivate the links.
Image matting evaluation results | Competition: Low resolution High resolution Error type: SAD MSE Gradient Connectivity |
Gradient error |
overall |
avg. small |
avg. large |
avg. user |
Troll (Strongly Transparent) Input |
Doll (Strongly Transparent) Input |
Donkey (Medium Transparent) Input |
Elephant (Medium Transparent) Input |
Plant (Little Transparent) Input |
Pineapple (Little Transparent) Input |
Plastic bag (Highly Transparent) Input |
Net (Highly Transparent) Input |
||||||||||||||||
rank | rank | rank | rank | small | large | user | small | large | user | small | large | user | small | large | user | small | large | user | small | large | user | small | large | user | small | large | user | |
LFPNet | 3.3 | 2.8 | 3.1 | 3.9 | 0.1 1 | 0.1 2 | 0.1 2 | 0.1 4 | 0.1 4 | 0.2 5 | 0.1 6 | 0.1 6 | 0.1 2 | 0.1 2 | 0.1 1 | 0.3 8 | 0.7 1 | 0.7 1 | 1.1 1 | 0.3 1 | 0.3 1 | 0.5 1 | 0.4 2 | 0.5 4 | 0.6 6 | 0.2 5 | 0.3 6 | 0.3 6 |
TMFNet | 4.3 | 2.9 | 4 | 6.1 | 0.1 2 | 0.1 1 | 0.2 7 | 0.1 2 | 0.1 11 | 0.2 10 | 0.1 2 | 0.1 2 | 0.2 9 | 0.1 1 | 0.1 2 | 0.3 5 | 0.7 2 | 0.9 3 | 1.5 6 | 0.3 2 | 0.4 3 | 0.7 6 | 0.5 6 | 0.6 7 | 0.5 4 | 0.3 6 | 0.3 3 | 0.2 2 |
TransMatting: Enhancing Transparent ... | 4.5 | 4.9 | 4.5 | 4.3 | 0.1 3 | 0.1 4 | 0.1 4 | 0.1 1 | 0.1 2 | 0.1 2 | 0.1 4 | 0.1 4 | 0.1 3 | 0.1 3 | 0.1 4 | 0.2 3 | 1 8 | 1.1 8 | 1.8 9 | 0.5 9 | 0.5 6 | 0.7 5 | 0.4 1 | 0.5 1 | 0.5 1 | 0.3 10 | 0.3 7 | 0.3 7 |
RMat | 4.6 | 4.6 | 3.6 | 5.6 | 0.1 4 | 0.1 3 | 0.1 1 | 0.1 3 | 0.1 1 | 0.2 15 | 0.1 1 | 0.1 1 | 0.1 1 | 0.1 6 | 0.1 3 | 0.3 6 | 0.8 3 | 0.8 2 | 1.3 2 | 0.4 4 | 0.4 4 | 0.7 3 | 0.6 9 | 0.6 10 | 0.7 12 | 0.3 7 | 0.3 5 | 0.3 5 |
IamAlpha | 6.2 | 7.9 | 6.8 | 3.9 | 0.1 6 | 0.1 6 | 0.1 3 | 0.1 7 | 0.1 3 | 0.1 1 | 0.2 13 | 0.2 11 | 0.2 8 | 0.2 12 | 0.2 10 | 0.2 1 | 0.9 5 | 1 6 | 1.4 5 | 0.6 14 | 0.7 12 | 0.9 10 | 0.5 5 | 0.5 5 | 0.5 2 | 0.2 1 | 0.2 1 | 0.2 1 |
SIM | 7.8 | 9.4 | 6.9 | 7 | 0.2 11 | 0.2 13 | 0.2 8 | 0.1 20 | 0.1 12 | 0.3 23 | 0.2 11 | 0.1 7 | 0.2 5 | 0.2 11 | 0.1 7 | 0.3 7 | 1 7 | 1 4 | 1.4 3 | 0.5 8 | 0.5 7 | 0.7 4 | 0.4 3 | 0.5 3 | 0.5 3 | 0.2 4 | 0.2 2 | 0.3 3 |
TIMI-Net | 7.8 | 8.8 | 8.4 | 6.3 | 0.2 12 | 0.2 12 | 0.2 9 | 0.1 17 | 0.1 15 | 0.1 4 | 0.1 8 | 0.1 8 | 0.2 6 | 0.2 13 | 0.1 8 | 0.3 9 | 1.1 10 | 1 7 | 1.4 4 | 0.3 3 | 0.4 2 | 0.5 2 | 0.4 4 | 0.5 2 | 0.6 7 | 0.2 3 | 0.4 13 | 0.3 9 |
HDMatt | 11.2 | 12.1 | 9.5 | 11.9 | 0.2 21 | 0.2 11 | 0.2 27 | 0.1 14 | 0.1 14 | 0.3 22 | 0.1 3 | 0.1 5 | 0.2 7 | 0.2 15 | 0.2 12 | 0.3 4 | 1.1 13 | 1.2 9 | 1.6 7 | 0.6 10 | 0.6 9 | 0.9 11 | 0.5 8 | 0.5 6 | 0.6 5 | 0.3 13 | 0.4 10 | 0.4 12 |
LiteMatting | 11.3 | 9.4 | 7.8 | 16.6 | 0.1 9 | 0.1 10 | 0.2 14 | 0.1 12 | 0.1 8 | 0.3 26 | 0.2 14 | 0.2 12 | 0.2 18 | 0.1 10 | 0.2 9 | 0.5 39 | 0.9 4 | 1 5 | 1.6 8 | 0.5 7 | 0.5 5 | 0.8 7 | 0.6 11 | 0.6 9 | 0.7 13 | 0.3 8 | 0.3 4 | 0.3 8 |
FGI Matting | 11.5 | 12.4 | 11.6 | 10.6 | 0.2 20 | 0.1 7 | 0.2 35 | 0.1 6 | 0.1 9 | 0.2 6 | 0.1 5 | 0.1 3 | 0.1 4 | 0.2 17 | 0.2 14 | 0.2 2 | 1.2 20 | 1.6 22 | 1.9 16 | 0.6 15 | 0.7 14 | 0.9 9 | 0.5 7 | 0.6 8 | 0.6 9 | 0.3 9 | 0.4 16 | 0.3 4 |
A2U Matting | 14 | 12.9 | 10.9 | 18.1 | 0.2 19 | 0.2 17 | 0.2 39 | 0.1 10 | 0.1 6 | 0.2 13 | 0.1 10 | 0.2 10 | 0.2 11 | 0.2 14 | 0.2 11 | 0.4 12 | 1.1 11 | 1.3 11 | 1.9 12 | 0.6 12 | 0.7 13 | 1.7 31 | 0.6 12 | 0.6 11 | 0.6 11 | 0.3 15 | 0.3 8 | 0.4 16 |
PIIAMatting | 14.2 | 9.3 | 15.4 | 18 | 0.1 8 | 0.2 18 | 0.2 12 | 0.1 9 | 0.1 5 | 0.2 11 | 0.1 7 | 0.2 17 | 0.2 17 | 0.1 5 | 0.4 26 | 0.6 52 | 0.9 6 | 1.5 16 | 1.9 15 | 0.4 6 | 0.7 10 | 1.2 18 | 1.4 31 | 0.7 13 | 0.6 8 | 0.2 2 | 0.4 18 | 0.3 11 |
AdaMatting | 15.1 | 10.6 | 12.9 | 21.8 | 0.2 16 | 0.2 19 | 0.2 36 | 0.1 5 | 0.1 7 | 0.4 47 | 0.2 18 | 0.2 15 | 0.2 14 | 0.1 4 | 0.1 5 | 0.3 11 | 1.1 9 | 1.4 13 | 2.3 24 | 0.4 5 | 0.6 8 | 0.9 8 | 0.9 17 | 1 19 | 0.9 20 | 0.3 11 | 0.4 17 | 0.4 14 |
LSA Matting | 15.3 | 14.4 | 15.1 | 16.4 | 0.1 5 | 0.1 5 | 0.1 5 | 0.2 25 | 0.3 45 | 0.2 16 | 0.2 15 | 0.2 16 | 0.2 12 | 0.1 8 | 0.1 6 | 0.4 16 | 1.2 19 | 1.4 12 | 1.8 11 | 0.6 13 | 0.7 11 | 1 12 | 0.9 18 | 0.9 17 | 1.1 49 | 0.3 12 | 0.4 9 | 0.3 10 |
GCA Matting | 16.1 | 16.3 | 14.9 | 17.3 | 0.1 10 | 0.1 8 | 0.2 16 | 0.1 19 | 0.1 17 | 0.3 24 | 0.2 17 | 0.2 19 | 0.2 19 | 0.2 19 | 0.2 15 | 0.3 10 | 1.3 21 | 1.6 21 | 1.9 13 | 0.7 16 | 0.8 15 | 1.4 24 | 0.6 10 | 0.7 12 | 0.6 10 | 0.4 18 | 0.4 12 | 0.4 22 |
Context-aware Matting | 17.3 | 18.8 | 18.4 | 14.6 | 0.2 13 | 0.2 16 | 0.2 6 | 0.1 15 | 0.2 19 | 0.2 9 | 0.2 20 | 0.2 20 | 0.2 13 | 0.2 22 | 0.4 23 | 0.4 23 | 1.4 25 | 1.5 14 | 1.8 10 | 0.8 18 | 1.3 22 | 1 13 | 1.1 21 | 1.1 22 | 0.9 28 | 0.4 16 | 0.4 11 | 0.4 15 |
SampleNet Matting | 17.3 | 12.8 | 15.4 | 23.9 | 0.1 7 | 0.1 9 | 0.2 34 | 0.1 8 | 0.1 18 | 0.2 8 | 0.2 22 | 0.3 21 | 0.3 21 | 0.1 9 | 0.2 13 | 0.5 28 | 1.1 14 | 1.5 15 | 2.7 37 | 0.6 11 | 0.9 16 | 1 14 | 0.8 14 | 0.9 16 | 0.9 32 | 0.4 17 | 0.4 15 | 0.4 17 |
IndexNet Matting | 21.4 | 19.9 | 20.3 | 24 | 0.2 15 | 0.2 15 | 0.2 17 | 0.1 16 | 0.1 13 | 0.3 18 | 0.2 12 | 0.2 18 | 0.2 10 | 0.2 16 | 0.2 16 | 0.4 14 | 1.7 34 | 1.9 32 | 2.5 30 | 1 22 | 1.1 20 | 1.3 21 | 1.1 23 | 1.2 26 | 1.2 55 | 0.4 21 | 0.5 22 | 0.5 27 |
ATNet Matting | 21.7 | 23.5 | 19.5 | 22.1 | 0.2 29 | 0.2 23 | 0.2 10 | 0.1 11 | 0.1 10 | 0.1 3 | 0.2 19 | 0.2 14 | 0.2 20 | 0.4 34 | 0.4 24 | 0.5 33 | 1.1 16 | 1.3 10 | 2.2 20 | 0.9 20 | 1.1 19 | 1.2 17 | 1.5 45 | 1.4 42 | 1.2 61 | 0.3 14 | 0.4 14 | 0.4 13 |
VDRN Matting | 22.5 | 24.8 | 20.4 | 22.4 | 0.2 18 | 0.2 14 | 0.2 11 | 0.1 18 | 0.1 16 | 0.3 30 | 0.2 16 | 0.2 13 | 0.2 16 | 0.4 36 | 0.3 21 | 0.5 25 | 1.4 27 | 1.6 20 | 1.9 14 | 0.8 19 | 0.9 18 | 1.1 15 | 0.9 16 | 0.9 18 | 0.7 14 | 0.7 48 | 0.7 43 | 0.7 54 |
Deep Matting | 26.3 | 23.3 | 23.1 | 32.5 | 0.4 62 | 0.4 59 | 0.5 62 | 0.2 21 | 0.2 29 | 0.2 7 | 0.1 9 | 0.1 9 | 0.2 15 | 0.2 21 | 0.2 18 | 0.6 51 | 1.3 23 | 1.5 19 | 2.4 28 | 0.8 17 | 0.9 17 | 1.3 23 | 0.7 13 | 0.8 14 | 1.1 46 | 0.4 20 | 0.5 20 | 0.5 28 |
AlphaGAN | 26.5 | 25.6 | 24.6 | 29.4 | 0.2 31 | 0.2 21 | 0.2 31 | 0.2 23 | 0.2 21 | 0.3 19 | 0.2 21 | 0.3 22 | 0.3 22 | 0.2 18 | 0.2 17 | 0.4 19 | 1.8 37 | 2.4 45 | 2.7 41 | 1.1 28 | 1.4 26 | 1.5 27 | 0.9 19 | 1.1 20 | 1 35 | 0.5 28 | 0.5 25 | 0.6 41 |
DCNN Matting | 27.9 | 31.6 | 27.9 | 24.1 | 0.2 25 | 0.2 22 | 0.2 19 | 0.2 37 | 0.3 37 | 0.3 21 | 0.3 25 | 0.4 29 | 0.3 24 | 0.3 28 | 0.4 22 | 0.4 13 | 1.5 29 | 1.5 18 | 2.1 17 | 1.1 26 | 1.3 24 | 1.5 25 | 1.5 47 | 1.4 41 | 1 36 | 0.6 36 | 0.6 30 | 0.5 38 |
CDI-Net | 29.5 | 27.5 | 25.8 | 35.3 | 0.3 48 | 0.2 31 | 0.3 45 | 0.1 13 | 0.2 20 | 0.2 12 | 0.3 23 | 0.3 23 | 0.3 23 | 0.3 24 | 0.3 20 | 0.6 47 | 1.8 38 | 1.9 36 | 2.4 27 | 1.4 35 | 1.6 30 | 2.4 47 | 0.8 15 | 0.9 15 | 0.9 23 | 0.5 24 | 0.6 31 | 0.9 58 |
Information-flow matting | 29.6 | 32.8 | 28.3 | 27.8 | 0.2 22 | 0.2 20 | 0.2 28 | 0.2 41 | 0.2 30 | 0.4 49 | 0.4 39 | 0.4 35 | 0.4 34 | 0.3 29 | 0.4 25 | 0.4 22 | 1.7 36 | 1.8 29 | 2.2 18 | 0.9 21 | 1.3 23 | 1.3 20 | 1.5 43 | 1.4 35 | 0.8 16 | 0.5 31 | 0.6 29 | 0.5 35 |
Three-layer graph matting | 31.5 | 27.6 | 29.8 | 37.3 | 0.2 14 | 0.2 24 | 0.2 32 | 0.2 31 | 0.2 25 | 0.4 52 | 0.3 24 | 0.4 25 | 0.3 25 | 0.1 7 | 0.3 19 | 0.5 32 | 1.6 31 | 1.7 24 | 2.7 39 | 1.1 29 | 1.9 37 | 2.4 49 | 1.6 58 | 1.6 58 | 1 45 | 0.5 27 | 0.6 26 | 0.4 24 |
Patch-based Matting | 32.5 | 29.5 | 32.6 | 35.5 | 0.2 37 | 0.2 29 | 0.2 23 | 0.2 26 | 0.3 47 | 0.3 42 | 0.4 27 | 0.4 31 | 0.4 33 | 0.3 27 | 0.5 32 | 0.6 48 | 1.2 17 | 1.5 17 | 2.2 19 | 1.3 34 | 2.1 39 | 1.9 38 | 1.2 24 | 1.2 25 | 0.9 34 | 0.6 44 | 0.6 41 | 0.6 47 |
Graph-based sparse matting | 32.6 | 30.4 | 32 | 35.4 | 0.2 38 | 0.3 33 | 0.2 18 | 0.2 29 | 0.2 26 | 0.4 51 | 0.4 33 | 0.4 30 | 0.4 35 | 0.2 20 | 0.5 27 | 0.4 21 | 1.9 43 | 1.9 33 | 2.8 43 | 1.1 27 | 1.8 33 | 1.8 34 | 1.4 34 | 1.5 50 | 1.2 52 | 0.4 19 | 0.5 24 | 0.5 29 |
KL-Divergence Based Sparse Sampling | 32.7 | 31.4 | 32.5 | 34.1 | 0.2 23 | 0.2 25 | 0.2 20 | 0.2 27 | 0.2 28 | 0.3 44 | 0.4 29 | 0.4 28 | 0.4 28 | 0.3 25 | 0.6 36 | 0.4 15 | 1.7 35 | 1.9 31 | 2.9 47 | 1.7 45 | 2.3 43 | 2.3 46 | 1.4 32 | 1.3 31 | 1 39 | 0.6 35 | 0.6 38 | 0.5 34 |
LNSP Matting | 33.3 | 30.9 | 34.3 | 34.9 | 0.2 24 | 0.3 35 | 0.2 13 | 0.2 22 | 0.2 22 | 0.3 40 | 0.4 28 | 0.5 43 | 0.4 31 | 0.3 23 | 0.5 29 | 0.6 45 | 1.8 40 | 1.9 34 | 2.7 42 | 1.2 30 | 1.6 31 | 2 42 | 1.4 41 | 1.3 32 | 0.9 27 | 0.6 39 | 0.8 48 | 0.6 39 |
Comprehensive sampling | 33.8 | 34 | 35.1 | 32.1 | 0.2 28 | 0.2 28 | 0.2 25 | 0.2 43 | 0.3 46 | 0.4 50 | 0.4 32 | 0.4 27 | 0.4 32 | 0.3 31 | 0.5 30 | 0.4 20 | 1.1 15 | 1.7 25 | 2.3 22 | 1.5 42 | 2.4 44 | 1.9 37 | 1.3 28 | 1.3 34 | 0.9 22 | 0.7 53 | 0.7 47 | 0.7 49 |
Three Stages Matting | 36 | 37.5 | 36.5 | 33.9 | 0.3 52 | 0.2 32 | 0.2 26 | 0.2 24 | 0.3 32 | 0.3 25 | 0.4 37 | 0.5 41 | 0.4 41 | 0.4 44 | 0.6 34 | 0.6 43 | 1.8 39 | 3.2 57 | 3 51 | 1 24 | 1.2 21 | 1.2 16 | 1.4 38 | 1.4 39 | 0.9 33 | 0.6 42 | 0.6 36 | 0.5 36 |
ATPM Matting | 36.1 | 37.4 | 40.9 | 30 | 0.2 36 | 0.3 42 | 0.2 21 | 0.2 36 | 0.2 23 | 0.3 45 | 0.4 35 | 0.5 44 | 0.4 30 | 0.4 35 | 0.7 44 | 0.4 24 | 2.1 47 | 2.3 44 | 2.6 35 | 1.2 31 | 1.8 34 | 1.5 26 | 1.4 30 | 1.4 40 | 0.8 17 | 0.7 49 | 0.9 56 | 0.6 42 |
Sparse coded matting | 36.3 | 38.3 | 36.5 | 34.3 | 0.3 54 | 0.4 51 | 0.3 55 | 0.3 48 | 0.3 41 | 0.3 39 | 0.4 48 | 0.5 37 | 0.4 46 | 0.3 32 | 0.5 28 | 0.4 17 | 1.4 26 | 1.7 28 | 2.3 25 | 1.1 25 | 1.6 27 | 1.3 22 | 1.5 48 | 1.6 57 | 1.1 50 | 0.5 25 | 0.5 23 | 0.4 20 |
SVR Matting | 36.9 | 39.6 | 39.3 | 31.9 | 0.3 55 | 0.4 58 | 0.3 46 | 0.3 54 | 0.3 40 | 0.3 34 | 0.4 36 | 0.5 42 | 0.4 29 | 0.4 37 | 0.6 40 | 0.5 29 | 1.5 28 | 2.1 39 | 2.5 31 | 1.2 32 | 1.6 29 | 1.5 28 | 1.3 25 | 1.2 24 | 0.8 18 | 0.7 50 | 0.7 42 | 0.6 40 |
CCM | 37.1 | 40.4 | 38.3 | 32.6 | 0.2 30 | 0.3 37 | 0.2 24 | 0.2 38 | 0.3 36 | 0.3 33 | 0.4 45 | 0.5 45 | 0.4 44 | 0.4 48 | 0.6 35 | 0.5 27 | 2.1 48 | 2.3 43 | 2.5 32 | 1.3 33 | 1.7 32 | 1.6 29 | 1.3 27 | 1.3 29 | 0.8 19 | 0.7 54 | 0.8 49 | 0.7 53 |
Segmentation-based matting | 37.2 | 41.6 | 36.3 | 33.6 | 0.2 32 | 0.3 34 | 0.2 29 | 0.2 34 | 0.2 24 | 0.3 17 | 0.4 41 | 0.5 50 | 0.4 39 | 0.4 38 | 0.6 33 | 0.5 30 | 2.1 49 | 1.7 27 | 2.5 33 | 1.9 48 | 2.2 41 | 2.5 50 | 2 65 | 1.7 60 | 1.2 53 | 0.5 26 | 0.5 21 | 0.4 18 |
Global Sampling Matting | 37.5 | 37.4 | 40.8 | 34.4 | 0.2 17 | 0.3 43 | 0.2 22 | 0.2 32 | 0.3 33 | 0.3 29 | 0.4 38 | 0.5 46 | 0.4 37 | 0.4 43 | 0.8 47 | 0.6 46 | 2.1 50 | 2.1 41 | 2.7 38 | 2 50 | 2.8 52 | 2.6 51 | 1.4 36 | 1.3 30 | 0.9 26 | 0.5 33 | 0.6 34 | 0.5 26 |
Improved color matting | 37.9 | 40.3 | 40.4 | 33 | 0.2 33 | 0.3 36 | 0.2 33 | 0.2 40 | 0.3 39 | 0.3 20 | 0.4 40 | 0.5 49 | 0.4 43 | 0.5 51 | 0.7 46 | 0.5 40 | 2.5 54 | 2.5 50 | 2.7 40 | 1.9 47 | 2.5 47 | 2.4 48 | 1.4 35 | 1.4 37 | 0.9 21 | 0.4 22 | 0.4 19 | 0.4 19 |
Shared Matting | 38.1 | 38.1 | 40.8 | 35.4 | 0.2 26 | 0.3 38 | 0.2 30 | 0.3 58 | 0.6 58 | 0.5 59 | 0.4 31 | 0.4 36 | 0.4 36 | 0.4 40 | 0.8 48 | 0.5 26 | 1.6 33 | 2 38 | 2.3 23 | 1.4 36 | 1.9 35 | 1.8 33 | 1.5 44 | 1.5 46 | 1.1 51 | 0.6 37 | 0.6 27 | 0.4 25 |
Comprehensive Weighted Color and Texture | 38.9 | 37.8 | 43.4 | 35.6 | 0.4 60 | 0.4 60 | 0.5 61 | 0.4 60 | 0.5 54 | 0.5 54 | 0.3 26 | 0.4 24 | 0.4 26 | 0.3 30 | 0.6 37 | 0.4 18 | 1.1 12 | 1.8 30 | 2.4 26 | 1 23 | 1.4 25 | 1.2 19 | 1.7 62 | 1.8 62 | 1.2 58 | 0.5 29 | 0.9 55 | 0.4 23 |
CSC Matting | 39.3 | 42.5 | 30.6 | 44.9 | 0.3 44 | 0.2 27 | 0.3 52 | 0.3 47 | 0.3 34 | 0.6 60 | 0.5 52 | 0.4 34 | 0.5 53 | 0.5 53 | 0.6 38 | 0.7 57 | 1.6 32 | 1.7 23 | 2.6 34 | 1.5 38 | 1.6 28 | 2 41 | 1.4 42 | 1.3 33 | 0.9 30 | 0.5 32 | 0.6 28 | 0.5 32 |
TSPS-RV Matting | 40.1 | 45.1 | 33.6 | 41.6 | 0.2 27 | 0.2 30 | 0.2 15 | 0.2 44 | 0.2 27 | 0.3 38 | 0.5 51 | 0.4 32 | 0.4 50 | 0.3 26 | 0.5 31 | 0.7 56 | 2.9 60 | 2.2 42 | 2.9 45 | 2.4 54 | 2.2 42 | 2.9 56 | 1.6 59 | 1.2 28 | 1 42 | 0.6 40 | 0.6 37 | 0.5 31 |
Global Sampling Matting (filter version) | 43.2 | 40.6 | 44.5 | 44.5 | 0.2 34 | 0.4 57 | 0.3 47 | 0.3 56 | 0.4 50 | 0.4 53 | 0.5 57 | 0.5 52 | 0.5 58 | 0.6 58 | 0.8 50 | 0.7 58 | 1.3 22 | 1.7 26 | 2.2 21 | 2 49 | 2.7 50 | 2.2 44 | 1.3 26 | 1.2 27 | 1.2 54 | 0.4 23 | 0.7 44 | 0.4 21 |
LocalSamplingAndKnnClassification | 43.5 | 44.6 | 44.3 | 41.5 | 0.2 42 | 0.3 41 | 0.2 38 | 0.2 28 | 0.3 31 | 0.3 28 | 0.4 30 | 0.4 33 | 0.4 27 | 0.4 45 | 0.7 43 | 0.5 41 | 2.9 59 | 3.1 56 | 3.9 59 | 2.5 55 | 3.2 57 | 2.9 55 | 1.6 52 | 1.5 47 | 1 40 | 0.7 46 | 0.7 46 | 0.6 44 |
LNCLM matting | 44.4 | 46.5 | 45.8 | 41 | 0.2 35 | 0.2 26 | 0.3 42 | 0.3 53 | 0.4 49 | 0.3 37 | 0.5 59 | 0.6 61 | 0.5 56 | 0.5 50 | 0.7 42 | 0.6 42 | 2.1 51 | 2.6 52 | 2.9 44 | 1.5 40 | 2.4 45 | 2 39 | 1.6 50 | 1.6 52 | 0.9 25 | 0.6 34 | 0.6 39 | 0.6 43 |
Weighted Color and Texture Matting | 45.2 | 43.8 | 45.1 | 46.6 | 0.2 41 | 0.3 44 | 0.2 37 | 0.3 55 | 0.4 51 | 0.5 56 | 0.4 34 | 0.4 26 | 0.4 38 | 0.5 52 | 1.1 56 | 0.7 59 | 1.9 44 | 2.4 46 | 3.3 55 | 1.5 41 | 2 38 | 1.9 36 | 1.4 40 | 1.4 43 | 1.1 47 | 0.6 43 | 0.9 57 | 0.6 45 |
SPS matting | 45.4 | 44.3 | 48.8 | 43.3 | 0.2 39 | 0.3 49 | 0.3 41 | 0.4 61 | 0.7 61 | 0.5 55 | 0.5 53 | 0.5 54 | 0.5 57 | 0.6 60 | 1.4 61 | 0.7 54 | 1.3 24 | 2 37 | 2.5 29 | 2.1 52 | 2.7 51 | 2.1 43 | 1 20 | 1.2 23 | 0.8 15 | 0.6 45 | 0.8 54 | 0.7 52 |
Iterative Transductive Matting | 46 | 47.8 | 45.8 | 44.5 | 0.3 47 | 0.4 52 | 0.3 43 | 0.2 45 | 0.6 57 | 0.5 58 | 0.5 61 | 0.5 38 | 0.4 51 | 0.3 33 | 1.1 55 | 0.5 38 | 2 46 | 2.1 40 | 3 50 | 1.8 46 | 2.5 48 | 2.3 45 | 1.6 57 | 1.4 36 | 1 38 | 0.7 47 | 0.6 40 | 0.5 33 |
SRLO Matting | 47.2 | 47.5 | 50.3 | 43.9 | 0.3 58 | 0.4 54 | 0.3 57 | 0.3 59 | 0.6 59 | 0.6 61 | 0.5 55 | 0.5 39 | 0.4 40 | 0.4 47 | 1.2 57 | 0.5 31 | 1.2 18 | 1.9 35 | 2.6 36 | 1.4 37 | 2.5 49 | 1.8 32 | 1.6 54 | 1.9 64 | 1 43 | 0.7 52 | 0.7 45 | 0.7 51 |
Improving Sampling Criterion | 47.3 | 47.9 | 48 | 46.1 | 0.3 46 | 0.4 50 | 0.3 50 | 0.3 57 | 0.4 52 | 0.4 46 | 0.5 56 | 0.5 57 | 0.5 59 | 0.9 64 | 1.4 62 | 0.9 62 | 2.2 52 | 2.7 53 | 2.9 46 | 2.6 56 | 3.2 56 | 2.6 52 | 1.1 22 | 1.1 21 | 0.9 24 | 0.5 30 | 0.6 33 | 0.5 30 |
Learning Based Matting | 48.1 | 46.5 | 50 | 47.8 | 0.3 51 | 0.3 46 | 0.3 53 | 0.2 35 | 0.3 35 | 0.3 27 | 0.4 47 | 0.5 55 | 0.4 49 | 0.4 41 | 0.6 39 | 0.5 37 | 1.9 42 | 2.5 49 | 3.2 54 | 3.1 63 | 3.6 62 | 3.9 63 | 1.4 33 | 1.6 54 | 1 37 | 0.9 60 | 1.1 60 | 1.2 62 |
LMSPIR | 48.3 | 47.1 | 49.1 | 48.6 | 0.3 57 | 0.4 55 | 0.3 56 | 0.3 52 | 0.7 60 | 0.5 57 | 0.4 50 | 0.5 40 | 0.4 48 | 0.4 39 | 1.2 58 | 0.5 35 | 1.5 30 | 2.5 48 | 3.2 53 | 1.6 43 | 2.4 46 | 1.9 35 | 1.6 55 | 1.6 51 | 1.2 57 | 0.7 51 | 0.6 35 | 0.6 48 |
Local Spline Regression (LSR) | 48.9 | 50.3 | 50 | 46.4 | 0.2 40 | 0.3 39 | 0.3 40 | 0.2 33 | 0.3 43 | 0.3 36 | 0.4 46 | 0.5 47 | 0.4 42 | 0.4 42 | 0.7 45 | 0.5 36 | 3.4 63 | 3.5 60 | 4.5 60 | 2.9 59 | 3 54 | 3.2 57 | 1.6 60 | 1.7 59 | 1 44 | 0.8 59 | 0.8 53 | 0.9 56 |
Closed-Form Matting | 48.9 | 49 | 49.6 | 48.1 | 0.3 45 | 0.3 45 | 0.3 49 | 0.2 30 | 0.3 38 | 0.3 32 | 0.4 43 | 0.5 53 | 0.4 45 | 0.4 46 | 0.6 41 | 0.6 44 | 3.3 62 | 3.6 61 | 4.6 61 | 2.9 62 | 3.5 60 | 3.7 62 | 1.5 49 | 1.5 48 | 0.9 31 | 0.7 55 | 0.8 51 | 1.2 61 |
KNN Matting | 49.1 | 52.5 | 49.5 | 45.4 | 0.3 53 | 0.3 47 | 0.3 44 | 0.3 49 | 0.3 42 | 0.3 41 | 0.6 64 | 0.6 60 | 0.6 64 | 0.5 55 | 1 52 | 0.6 49 | 2.8 57 | 2.8 54 | 3 49 | 1.5 39 | 1.9 36 | 1.6 30 | 1.5 46 | 1.5 44 | 0.9 29 | 0.8 57 | 1.2 61 | 0.9 57 |
Robust Matting | 50.3 | 47 | 50.6 | 53.1 | 0.3 49 | 0.4 53 | 0.3 54 | 0.5 62 | 0.9 62 | 0.6 62 | 0.4 44 | 0.5 48 | 0.5 54 | 0.5 54 | 0.8 49 | 0.7 53 | 1.8 41 | 2.6 51 | 3.1 52 | 2 51 | 3.1 55 | 2.7 53 | 1.4 37 | 1.6 55 | 1.2 60 | 0.6 38 | 0.6 32 | 0.5 37 |
Large Kernel Matting | 51 | 51.5 | 52.1 | 49.4 | 0.3 50 | 0.3 48 | 0.3 51 | 0.2 39 | 0.3 44 | 0.2 14 | 0.5 54 | 0.6 59 | 0.5 55 | 0.5 56 | 1.3 59 | 0.7 55 | 2.9 58 | 3.3 59 | 3.7 56 | 2.6 58 | 3.3 58 | 3.4 61 | 1.4 39 | 1.4 38 | 1.1 48 | 0.8 58 | 0.8 52 | 0.8 55 |
High-res matting | 51.7 | 49.4 | 55.8 | 50 | 0.2 43 | 0.3 40 | 0.3 48 | 0.2 46 | 0.5 55 | 0.4 48 | 0.4 42 | 0.5 51 | 0.4 47 | 0.4 49 | 1.3 60 | 0.5 34 | 2.3 53 | 2.9 55 | 3.7 58 | 2.6 57 | 3.8 63 | 3.3 60 | 1.8 64 | 1.8 63 | 1.2 59 | 0.6 41 | 1.1 59 | 0.6 46 |
Shared Matting (Real Time) | 53.4 | 53.9 | 53.1 | 53.1 | 0.3 59 | 0.5 62 | 0.4 60 | 0.6 63 | 0.9 63 | 0.7 63 | 0.4 49 | 0.5 56 | 0.4 52 | 0.6 59 | 1.1 54 | 0.6 50 | 2 45 | 2.5 47 | 2.9 48 | 1.7 44 | 2.1 40 | 2 40 | 1.6 56 | 1.6 53 | 1.3 62 | 0.8 56 | 0.8 50 | 0.7 50 |
Cell-based matting Laplacian | 56 | 57 | 56.1 | 54.9 | 0.3 56 | 0.4 56 | 0.3 58 | 0.3 51 | 0.6 56 | 0.3 43 | 0.6 62 | 0.6 62 | 0.6 62 | 0.6 57 | 0.9 51 | 0.7 60 | 2.7 56 | 3.2 58 | 3.7 57 | 2.9 60 | 3.4 59 | 3.2 59 | 1.6 53 | 1.5 49 | 1 41 | 0.9 61 | 1 58 | 0.9 59 |
Transfusive Weights | 59.2 | 57.3 | 60.5 | 59.8 | 0.5 63 | 0.5 64 | 0.5 63 | 0.2 42 | 0.4 48 | 0.3 31 | 0.5 60 | 0.6 63 | 0.6 63 | 0.9 63 | 1.5 65 | 1 63 | 4.9 68 | 5.8 66 | 7 68 | 4.5 68 | 5.6 68 | 5.3 68 | 1.3 29 | 1.5 45 | 1.2 56 | 1.5 65 | 1.6 65 | 1.8 66 |
Random Walk Matting | 60.5 | 61.4 | 60.3 | 59.8 | 0.4 61 | 0.5 61 | 0.4 59 | 0.3 50 | 0.4 53 | 0.3 35 | 0.5 58 | 0.6 58 | 0.5 61 | 0.7 61 | 1 53 | 0.8 61 | 4.8 67 | 6.4 68 | 6.7 67 | 4.4 67 | 5 67 | 4.7 67 | 1.7 61 | 1.6 56 | 1.3 63 | 1.7 66 | 1.9 66 | 1.7 65 |
Iterative BP Matting | 62.3 | 60.5 | 62.9 | 63.6 | 0.5 64 | 0.5 63 | 0.6 64 | 0.7 64 | 1.2 65 | 1.2 65 | 0.6 65 | 0.8 64 | 0.6 65 | 0.7 62 | 1.5 64 | 1.1 64 | 2.7 55 | 3.6 62 | 4.9 63 | 2.9 61 | 3.6 61 | 3.2 58 | 1.6 51 | 1.8 61 | 2.9 67 | 1.1 62 | 1.3 63 | 1.2 63 |
Geodesic Matting | 64.5 | 64.8 | 64.4 | 64.4 | 1.4 67 | 3.2 69 | 1.8 68 | 1.2 65 | 1.1 64 | 0.9 64 | 1.2 68 | 1.5 69 | 1 68 | 1.9 68 | 2.1 67 | 1.7 68 | 4.1 66 | 4.6 63 | 4.6 62 | 2.2 53 | 3 53 | 2.9 54 | 3.5 67 | 3.3 66 | 2 64 | 1.3 64 | 1.4 64 | 1.8 67 |
Improved Bayesian | 64.5 | 64.6 | 64.8 | 64.1 | 1.6 68 | 1.3 65 | 0.9 65 | 1.4 66 | 1.4 66 | 1.6 67 | 0.6 63 | 1.1 67 | 0.5 60 | 1 65 | 1.4 63 | 1.5 67 | 3.3 61 | 5 65 | 5.1 64 | 3.6 64 | 4.4 65 | 4.1 65 | 1.7 63 | 2 65 | 2.2 65 | 2.2 67 | 1.2 62 | 1.1 60 |
Bayesian Matting | 66.2 | 65.8 | 66.9 | 66 | 1.3 66 | 1.5 67 | 1.2 67 | 1.8 67 | 2 67 | 1.6 66 | 0.9 67 | 1.1 66 | 1 67 | 1.3 67 | 2.5 68 | 1.2 66 | 3.6 64 | 5.8 67 | 6.6 66 | 4 66 | 4.7 66 | 4.1 66 | 3.1 66 | 3.6 67 | 2.8 66 | 1.2 63 | 2 67 | 1.3 64 |
Easy Matting | 66.3 | 66.4 | 66.3 | 66.3 | 1.1 65 | 1.3 66 | 1 66 | 2.5 68 | 3 68 | 2.8 68 | 0.7 66 | 0.8 65 | 0.7 66 | 1.1 66 | 2.1 66 | 1.1 65 | 4.1 65 | 4.6 64 | 6.3 65 | 3.7 65 | 4.1 64 | 4.1 64 | 3.6 68 | 4.5 69 | 3.4 68 | 2.9 68 | 3.1 68 | 3.3 68 |
Poisson Matting | 68.9 | 69 | 68.6 | 69 | 2.3 69 | 2 68 | 2.6 69 | 3.4 69 | 3.6 69 | 3.4 69 | 1.5 69 | 1.4 68 | 1.3 69 | 3.1 69 | 3 69 | 3.1 69 | 6.6 69 | 7.9 69 | 9.9 69 | 6.7 69 | 7.4 69 | 6.6 69 | 3.8 69 | 4.2 68 | 4.5 69 | 4.1 69 | 5.3 69 | 4.3 69 |
Troll - Input image Drag the window to change the zoom. |
References
Method | Reference and notes | Implementation details |
Closed-Form Matting | A. Levin, D. Lischinski, Y. Weiss, A Closed Form Solution to Natural Image Matting, CVPR, 2006 | Matlab implementation on a Intel Core2 Quad with 2.4 GHZ |
Bayesian Matting | Y.Y. Chuang, B. Curless, D. Salesin, R. Szeliski, A Bayesian Approach to Digital Matting, CVPR, 2001 | C++ implementation on a Intel Core2 Quad with 2.4 GHZ |
Poisson Matting | J. Sun, J. Jia, C.K. Tang, H.Y. Shum, Poisson matting, SIGGRAPH, 2004 | Matlab implementation on a Intel Core2 Quad with 2.4 GHZ |
Easy Matting | Y. Guan, W. Cheny, X. Liang, Z. Ding, Q. Peng, Easy Matting: A Stroke Based Approach for Continuous Image Matting, Eurographics, 2006 | C++ implementation on a Intel Core2 Quad with 2.4 GHZ |
Random Walk Matting | L. Grady, T. Schiwietz, S. Aharon, Random Walks For Interactive Alpha-Matting, VIIP, 2005 | Matlab/C++ implementation on a Intel Core2 Quad with 2.4 GHZ |
Robust Matting | J. Wang, M. Cohen, Optimized Color Sampling for Robust Matting, CVPR, 2007 | C++ implementation on a Intel Core2 Quad with 2.4 GHZ |
Geodesic Matting | Xue Bai, Guillermo Sapiro, A geodesic framework for fast interactive image and video segmentation and matting, ICCV 2007 | C++ implementation on a Intel Core2 Duo with 2.53 GHZ |
Iterative BP Matting | Jue Wang, Michael Cohen, Jue Wang and Michael F. Cohen. An iterative optimization approach for unified image segmentation and matting. ICCV 2005. | c++ implementation on a Intel Core2 Quad with 3 GHZ |
Improved color matting | C. Rhemann, C. Rother, M. Gelautz, Improving Color Modeling for Alpha Matting. BMVC, 2008 | Matlab implementation on a Intel Core2 Duo with 2.4 GHZ |
High-res matting | C. Rhemann, C. Rother, A. Rav-Acha, M. Gelautz, T. Sharp, High ResolutionMatting via Interactive Trimap Segmentation. CVPR, 2008 | Matlab/C++ implementation on a Intel Core2 Duo with 2.4 GHZ |
Large Kernel Matting | Kaiming He, Jian Sun, and Xiaoou Tang, Fast Matting using Large Kernel Matting Laplacian Matrices, CVPR 2010 | C++ implementation on a Intel Core Duo with 2 GHZ |
Segmentation-based matting | Christoph Rhemann, Carsten Rother, Pushmeet Kohli, Margrit Gelautz, A Spatially Varying PSF-based Prior for Alpha Matting, CVPR 2010 | Matlab/C++ implementation on a Intel Core2 Quad with 2.39 GHZ |
Shared Matting | Eduardo S. L. Gastal and Manuel M. Oliveira, Shared Sampling for Real-Time Alpha Matting, Eurographics, 2010 | C++/GLSL implementation on a Core 2 Quad with 2.8 GHZ |
Shared Matting (Real Time) | Eduardo S. L. Gastal and Manuel M. Oliveira, Shared Sampling for Real-Time Alpha Matting, Eurographics, 2010 | C++/GLSL implementation on a Core 2 Quad with 2.8 GHZ |
Learning Based Matting | Yuanjie Zheng, Chandra Kambhamettu, Yuanjie Zheng, Chandra Kambhamettu. Learning Based Digital Matting. ICCV 2009. SOURE CODE | Matlab/C++ implementation on a Intel Core2 Duo with 2.53 GHZ |
LMSPIR | Bei He, Guijin Wang, Zhiwei Ruan, Xuanwu Yin, Xiaokang Pei, Xinggang Lin, Local Matting based on Sample-pair Propagation and Iterative Refinement, ICIP 2012 | C++ implementation on a Intel Core2 Dual with 2 GHZ |
SVR Matting | Zhanpeng Zhang, Qingsong Zhu, Yaoqin Xie, Learning Based Alpha Matting using Support Vector Regression, ICIP 2012 | Matlab implementation on a Intel Pentium Dual-Core with 3 GHZ |
Cell-based matting Laplacian | Chen-Yu Tseng and Sheng-Jyh Wang, A cell-based matting Laplacian for contrast enhancement, ICIP 2012 | C++ implementation on a Intel Core i3 with 3 GHZ |
Global Sampling Matting | Kaiming He, Christoph Rhemann, Carsten Rother, Xiaoou Tang, and Jian Sun, A Global Sampling Method for Alpha Matting, CVPR 2011 | C++ implementation on a Intel Core2 with 2 GHZ |
Global Sampling Matting (filter version) | Kaiming He, Christoph Rhemann, Carsten Rother, Xiaoou Tang, and Jian Sun, A Global Sampling Method for Alpha Matting, CVPR 2011 (using the guided filter in "Guided Image Filtering", ECCV 2010, by Kaiming He, Jian Sun, and Xiaoou Tang ) | C++ implementation on a Intel Core2 with 2 GHZ |
Local Spline Regression (LSR) | Shiming Xiang, Feiping Nie, Changshui Zhang, Semi-Supervised Classification via Local Spline Regression. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, no. 11, pp. 2039-2053, 2010 | C++ implementation on a Intel Core2 with 3 GHZ |
Weighted Color and Texture Matting | E.Shahrian and D.Rajan, Weighted Color and Texture Sample Selection for Image Matting , CVPR 2012. | matlab implementation on a Intel(R) Xeon(R) with 2.93 GHZ |
KNN Matting | Qifeng Chen, Dingzeyu Li, Chi-Keung Tang, KNN Matting, CVPR 2012 | Matlab implementation on a Intel Core 2 Duo with 2.13 GHZ |
SRLO Matting | Bei He, Guijin Wang, Xuanwu Yin, Bo Liu, Chenbo Shi, Xinggang Lin, High-accuracy and Quick Matting based on Sample-pair Refinement and Local Optimization, IEICE trans. on Information & Systems, 2013 | C++ implementation on a Intel Core2 Dual with 2 GHZ |
LNSP Matting | Xiaowu Chen, Dongqing Zou, Ping Tan, Image Matting with Local and Nonlocal Smooth Priors, CVPR 2013 | matlab implementation on a intel core2 with 2.2 GHZ |
CCM | Yongfang Shi, Au, O.C., Jiahao Pang, Tang, K., Wenxiu Sun, Hong Zhang, Wenjing Zhu, and Luheng Jia, Color Clustering Matting, ICME2013. | Matlab implementation on a Intel Core i7 with 2.8 GHZ |
Iterative Transductive Matting | Bei He, Guijin Wang, Chenbo Shi, Xuanwu Yin, Bo Liu, Xinggang Lin, Iterative Transductive Matting, ICIP 2013 | Matlab implementation on a Intel Core2 Dual with 2.2 GHZ |
Improving Sampling Criterion | Jun Cheng, Zhenjiang Miao, Improving Sampling Criterion for Alpha Matting, RACVPR2013 in Conjunction with ACPR2013 | C++ implementation on a Core i5 with 2.5 GHZ |
Transfusive Weights | Kaan Yucer, Alexander Sorkine-Hornung, and Olga Sorkine-Hornung, Transfusive Weights for Content-Aware Image Manipulation, VMV2013 | Matlab implementation on a Quad-Core Intel Xeon with 3.2 GHZ |
Comprehensive sampling | E.Shahrian, D.Rajan, B.Price, S.Cohen, Improving Image Matting using Comprehensive Sampling Sets, CVPR 2013. | Matlab implementation on a Intel Xeon with 2.4 GHZ |
Comprehensive Weighted Color and Texture | E.Shahrian , D.Rajan, Weighted Color and Texture Sample Selection for Image Matting, IEEE Transaction on Image Processing , Volume:PP, Issue: 99 , 2013. | Matlab implementation on a Intel i7 with 3.5 GHZ |
SPS matting | Ahmad Al-Kabbany and Eric Dubois, "Improved global-sampling matting using sequential pair-selection strategy", In Proc. Visual Information Processing andCommunication V(SPIE), San Francisco, February 2014. | Matlab implementation on a Intel Core2 Quad with 2.66 GHZ |
Improved Bayesian | Wenshuang Tan, Automatic Matting of Identification Photos, CAD/Graphics, 2013 | C++ implementation on a Intel Core(TM)i7-2600 with 3.4 GHZ |
Sparse coded matting | Jubin Johnson, Deepu Rajan, Hisham Cholakkal, Sparse Codes as Alpha Mattes, BMVC 2014. | Matlab implementation on a Intel Xeon with 3.2 GHZ |
LNCLM matting | B.-K. Kim, M. Jin, W.-J Song, Local and Nonlocal Color Line Models for Image Matting, IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, vol. E97-A. no.8, pp. 1814-1819, Aug. 2014. | MATLAB implementation on a Intel Core i5-3570 with 3.4 GHZ |
Graph-based sparse matting | Jubin Johnson, Ehsan Shahrian Varnousfaderani, Hisham Cholakkal, and Deepu Rajan, Sparse Coding for Alpha Matting, IEEE Transactions on Image Processing, Volume: PP, Issue: 99, 2016. | Matlab implementation on a Intel Xeon with 3.2 GHZ |
KL-Divergence Based Sparse Sampling | Levent Karacan, Aykut Erdem and Erkut Erdem, Image Matting with KL-Divergence Based Sparse Sampling,IEEE International Conference on Computer Vision(ICCV) 2015) | Matlab implementation on a Intel® Xeon(R) CPU E5-2620 with 2 GHZ |
LocalSamplingAndKnnClassification | Xiao Chen, Fazhi He, A Sampling-Propagation Matting Method Based on the Sample Validity and KNN Classification Labeling, Journal of Computer-Aided Design and Computer Graphics (CADCG), vol. 28(12), pp. 2186-2194, 2016. | matlab implementation on a intel i3 2120 with 3.3 GHZ |
DCNN Matting | Donghyeon Cho, Yu-Wing Tai, Inso Kweon, Natural Image Matting using Deep Convolutional Neural Networks. ECCV 2016 | matlab implementation on a Intel Core i7 with 3.4 GHZ |
CSC Matting | Xiaoxue Feng, Xiaohui Liang, Zili Zhang, A Cluster Sampling Method for Image Matting via Sparse Coding. ECCV 2016 | Matlab implementation on a Intel(R) Core(TM) i5-3470 with 3.2 GHZ |
Patch-based Matting | Guangying Cao, Jianwei Li, Zhiqiang He, Xiaowu Chen, Divide and Conquer: A Self-Adaptive Approach for High-Resolution Image Matting. International Conference on Virtual Reality and Visualization (ICVRV 2016) | matlab implementation on a i7 with 3.6 GHZ |
TSPS-RV Matting | Ahmad Al-Kabbany and Eric Dubois, Matting with Sequential Pair Selection Using Graph Transduction. The 21st International Symposium on Vision, Modeling, and Visualization (VMV 2016) | Matlab implementation on a Intel Core2 Quad with 2.66 GHZ |
Deep Matting | Ning Xu, Brian Price, Scott Cohen and Thomas Huang, Deep Image Matting, CVPR 2017 | Matlab implementation on a I7 with 2.7 GHZ |
Information-flow matting | Yagiz Aksoy, Tunc Ozan Aydin and Marc Pollefeys, Designing Effective Inter-Pixel Information Flow for Natural Image Matting, CVPR 2017 | Matlab implementation on a Intel Xeon with 3.5 GHZ |
ATPM Matting | Xiangyu Zhu, Ping Wang, Zhenghai Huang, Adaptive Propagation Matting Based on Transparency of Image, Multimedia Tools and Applications, vol. 77, pp. 9089-19112, 2018, doi: 10.1007/s11042-017-5357-7 | matlab implementation on a Intel Xeon with 2.4 GHZ |
Three-layer graph matting | Chao Li, Ping Wang, Xiangyu Zhu, Huali Pi, Three-layer graph framework with the sumD feature for alpha matting. Computer Vision and Image Understanding, vol. 162, pp. 34-45, 2017 | c++, octave implementation on a Intel Xeon with 3.2 GHZ |
Three Stages Matting | Xiao Chen, A Three-Stage Matting Method, IEEE Access, 5(99):27732-27739, 2017 | matlab implementation on a i7 6700k with 4 GHZ |
AlphaGAN | Sebastian Lutz, Konstantinos Amplianitis, Aljosa Smolic, AlphaGAN: Generative Adversarial Networks for Natural Image Matting, BMVC 2018 | Python implementation on a Intel i7-6700 with 3.4 GHZ |
SampleNet Matting | Jingwei Tang, Yagiz Aksoy, Cengiz Oztireli, Markus Gross, Tunc Ozan Aydin, Learning-based Sampling for Natural Image Matting, CVPR 2019 | Python implementation on a Intel Core i7-7700K with 4.7 GHZ |
VDRN Matting | Huan Tang, Yujie Huang, Ming'e Jing, Yibo Fan, Xiaoyang Zeng, Very deep residual network for image matting, IEEE ICIP 2019 | python implementation on a intel Xeon with 2.2 GHZ |
AdaMatting | Shaofan Cai, Xiaoshuai Zhang, Haoqiang Fan, Haibin Huang, Jiangyu Liu, Jiaming Liu, Jiaying Liu, Jue Wang, and Jian Sun, Disentangled Image Matting, ICCV 2019 | Python implementation on a Intel Core i7-7700K with 4.7 GHZ with 2.2 GHZ |
IndexNet Matting | Hao Lu, Yutong Dai, Chunhua Shen, Songcen Xu, Indices Matter: Learning to Index for Deep Image Matting, ICCV 2019 | Python (PyTorch) implementation on a Intel i7-8700, GTX1070 with 3.2 GHZ |
Context-aware Matting | Qiqi Hou, Feng Liu, Context-aware Image Matting for Simultaneous Foreground and Alpha Estimation. ICCV 2019 | Python, Tensorflow implementation on a 1080 Ti with 2.2 GHZ |
GCA Matting | Yaoyi Li, Hongtao Lu, Natural Image Matting via Guided Contextual Attention, AAAI 2020 | python implementation on a Intel(R) Xeon(R) CPU E5-2640, GeForce RTX 2080 Ti with 2.6 GHZ |
ATNet Matting | F. Zhou, Y. Tian, Z. Qi, Attention Transfer Network For Nature Image Matting, IEEE Transactions on Circuits and Systems for Video Technology, doi: 10.1109/TCSVT.2020.3024213 | python implementation on a 1080Ti with 2.2 GHZ |
PIIAMatting | Yuhao Liu, Jiake Xie, Yu Qiao, Yong Tang, Xin Yang, Prior-Induced Information Alignment for Image Matting, IEEE Transactions on Multimedia, doi: 10.1109/TMM.2021.3087007 | Python implementation on a GTX 2080ti with 2.6 GHZ |
SIM | Yanan Sun, Chi-Keung Tang, Yu-Wing Tai, Semantic Image Matting, CVPR 2021 | python implementation on a GTX 2080ti with 2.6 GHZ |
HDMatt | Haichao Yu, Ning Xu, Zilong Huang, Yuqian Zhou, Humphrey Shi, High-Resolution Deep Image Matting, AAAI 2021 | Python implementation on a Tesla V100 with 3.2 GHZ |
TIMI-Net | Yuhao Liu, Jiake Xie, Xiao Shi, Yu Qiao, Yujie Huang, Yong Tang, Xin Yang, Tripartite Information Mining and Integration for Image Matting, ICCV 2021 | python implementation on a tesla v100 with 3.5 GHZ |
A2U Matting | Yutong Dai, Hao Lu, Chunhua Shen, Learning Affinity-Aware Upsampling for Deep Image Matting, CVPR 2021 | Python implementation on a GTX 1080 Ti with 3.2 GHZ |
LFPNet | Qinglin Liu, Haozhe Xie, Shengping Zhang, Bineng Zhong, Rongrong Ji, Long-Range Feature Propagating for Natural Image Matting, ACM MM 2021 | Python implementation on a Nvidia GTX 1080Ti with 1.5 GHZ |
IamAlpha | Avinav Goel, Manoj Kumar, Pavan Sudheendra, IamAlpha: Instant and Adaptive Mobile Network for Alpha Matting, BMVC 2021 | Python implementation on a Nvidia Tesla P40 with 3.1 GHZ |
TMFNet | anonymous, Trimap-guided Feature Mining and Fusion Network for Natural Image Matting, submission to CVIU, 2022 | python implementation on a Tesla V100 with 3.5 GHZ |
FGI Matting | Hang Cheng, Shugong Xu, Xiufeng Jiang, Rongrong Wang, Deep Image Matting with Flexible Guidance Input, BMVC 2021 | python implementation on a RTX 2080 Ti with 2.6 GHZ |
LSA Matting | Rui Wang, Jun Xie, Jiacheng Han, Dezhen Qi, Improving Deep Image Matting via Local Smoothness Assumption, IEEE ICME 2022 | Python implementation on a Intel Xeon with 2.3 GHZ |
RMat | Yutong Dai, Brian Price, He Zhang, Chunhua Shen, Boosting Robustness of Image Matting with Context Assembling and Strong Data Augmentation, CVPR 2022 | python implementation on a Tesla V100 with 3.2 GHZ |
TransMatting: Enhancing Transparent ... | Huanqia Cai, Fanglei Xue, Lele Xu, Lili Guo, TransMatting: Enhancing Transparent Objects Matting with Transformers, ECCV 2022, accepted | Python implementation on a Intel(R) Xeon(R) Silver 4210R with 2.4 GHZ |
LiteMatting | anonymous, Lightweight Image Matting via Efficient Non-Local Guidance, anonymous submission 2022 | python implementation on a Intel with 3.8 GHZ |
CDI-Net | Zhiwei Ma, Guilin Yao, submission to Journal of Visual Communication and Image Representation, 2022 | Python implementation on a AMD Ryzen 9 5900HX with 3.3 GHZ |