Word Recognition on LRW (test)
98Correct RateA + V (End-to-End)
Evaluation Results
| Method | Links | |
|---|---|---|
| A + V (End-to-End)Stream=Audiovisual, Model Type=End-to-End, Mouth ROI=Fixed2018.02 | 98 | |
| A (End-to-End)Stream=Audio, Model Type=End-to-End, Input=Raw waveform2018.02 | 97.7 | |
| A (MFCC)Stream=Audio, Model Type=2-layer BGRU, Features=MFCC2018.02 | 97.7 | |
| Multi-Grained Spatio-temporal Network2019.08 | 83.34 | |
| V [13]*Stream=Video, Model Type=End-to-End, Mouth ROI=Tracked facial landmarks2018.02 | 83 | |
| Stafylakis172019.08 | 83 | |
| V (End-to-End)Stream=Video, Model Type=End-to-End, Mouth ROI=Fixed2018.02 | 82 | |
| Petridis18mode=end-to-end2019.08 | 82 | |
| Petridis18reproduction=true2019.08 | 81.7 | |
| V [15]Stream=Video2018.02 | 76.2 | |
| Chung172019.08 | 76.2 | |
| Chung182019.08 | 71.5 | |
| V [19]Stream=Video2018.02 | 61.1 |