Deep Video Compression
Our Test Sequence
The Common Test Conditions(CTC) of the test sequence(6 sequences) we used are as follows.
- Chroma Format: RGB 4:4:4
- Input bit-depth: 8
- Resolution: 768x768
- HEVC & VVC QP: 22, 27, 32, 37
- Compress AI Quality: 1, 2, 3, 4, 5, 6
Sensor-generated Sequence
| Input bit-depth | Frame rate | Test sequence name | frame count |
|---|---|---|---|
| 8 | 50 | CrowdRun | 100 |
| 8 | 50 | DucksTakeOff | 100 |
| 8 | 50 | OldTownCross | 100 |
| 8 | 50 | Parkjoy | 100 |
Computer-generated Sequence
| Input bit-depth | Frame rate | Test sequence name | frame count |
|---|---|---|---|
| 8 | 60 | ArenaOfValor | 120 |
| 8 | 24 | GlassHalf | 48 |
Reference Software
HEVC (HM 16.8): Download HM
VVC (VTM 12.1): Download VTM, Download Documents
Compress AI (bmshj2018-hyperprior, mbt2018, cheng2020-anchor): compressAI github
Our Proposed Codec
Step
Code explanation
The Directory Structure
.
+-- Our Encoder
+-- train
+-- test
+-- PostProcessing
+-- test
+-- PreProcessing
+-- test