Research Article
Adaptive CU Split Decision Based on Deep Learning and Multifeature Fusion for H.266/VVC
Table 1
The prediction accuracy of adaptive CNN under different parameter setting.
| (%) | Number of parameters | Number of inputs | Conv | Kernel | FCL | NB2 and NB4 | NB1, NB2, NB3, and NB4 |
| Default parameter | 2 | 6 | 2 | 73.43 | 76.82 |
| Comparison of Conv | 3 | 6 | 2 | 72.88 | 74.94 | 4 | 6 | 2 | 71.67 | 73.73 |
| Comparison of kernel | 2 | 8 | 2 | 80.55 | 82.76 | 2 | 10 | 2 | 88.96 | 91.67 |
| Comparison of FCL | 2 | 6 | 3 | 70.59 | 73.80 | 2 | 6 | 4 | 68.81 | 69.80 |
|
|