From: Variational approach for capsule video frame interpolation
Method | Seq1 | Seq2 | Seq3 | Seq4 | Avg. | |||||
---|---|---|---|---|---|---|---|---|---|---|
Frames ratio | 90% | 100% | 90% | 100% | 90% | 100% | 90% | 100% | 90% | 100% |
TV-CLG1 | 3.34 | 8.26 | 18.59 | 19.23 | 3.71 | 6.53 | 6.79 | 8.65 | 9.28 | 10.44 |
SFlow2 | 3.41 | 7.84 | 18 | 18.63 | 3.86 | 6.40 | 6.02 | 7.48 | 7.82 | 9.87 |
Frame averaging | 43.63 | 47.28 | 68.12 | 66.47 | 55.43 | 54.91 | 59.85 | 59.67 | 57.78 | 57.52 |
TV-L13 | 4.05 | 11.38 | 20.20 | 20.49 | 4.50 | 8.21 | 7.73 | 9.74 | 9.13 | 12.13 |
Proposed method without texture | 3.97 | 6.82 | 19.82 | 20.55 | 4.06 | 8.03 | 6.47 | 8.46 | 8.55 | 10.89 |
Proposed method with texture | 2.54 | 5.69 | 18.92 | 19.56 | 3.51 | 6.99 | 5.99 | 7.91 | 7.78 | 9.98 |