Research Article
Adaptive CU Split Decision Based on Deep Learning and Multifeature Fusion for H.266/VVC
Table 4
The coding performance of the proposed algorithm compared with
.
| Test sequence | The proposed method | BD-rate (%) | TS (%) |
| Class A1 | Tango2 | 0.78 | 38.87 | FoodMarket4 | 0.82 | 39.96 | Campfire | 0.89 | 41.55 |
| Class A2 | Catrobot1 | 0.92 | 40.14 | DaylightRoad2 | 0.85 | 39.58 | ParkRunning3 | 0.81 | 38.22 |
| Class B | Kimono | 0.78 | 37.51 | ParkScene | 0.61 | 39.56 | BQTerrace | 0.76 | 41.79 |
| Class C | PartyScene | 0.37 | 36.73 | RaceHorsesC | 0.24 | 30.68 | BasketballDrill | 1.25 | 39.21 |
| Class D | BlowingBubbles | 0.83 | 40.87 | RaceHorses | 0.56 | 36.51 | BQSquare | 0.58 | 36.67 |
| Class E | Johnny | 1.56 | 43.78 | FourPeople | 1.34 | 46.51 | KristenAndSara | 1.57 | 40.85 |
| ā | Average | 0.86 | 39.39 |
|
|