Lip Reading on LRW 1.0 (test)
94.8Top-1 AccuracyTranscription Alignment
Evaluation Results
| Method | Links | |
|---|---|---|
| Transcription AlignmentTemporal Model=Transformer, Video Hours=157, Word Boundary (WB)=true2024.06 | 94.8 | |
| DC-TCN (Ensemble)Word Boundary=true, Pre-training=LRS2&3+AVS, Ensemble=true2022.09 | 94.1 | |
| DC-TCN (Ensemble)Word Boundary=true, Pre-training=LiRA(LRS3), Ensemble=true2022.09 | 93.6 | |
| DC-TCN (Ensemble)Word Boundary=true, Pre-training=Scratch, Ensemble=true2022.09 | 93.4 | |
| Transcription AlignmentTemporal Model=Transformer, Video Hours=157, Word Boundary (WB)=false2024.06 | 93.1 | |
| DC-TCNWord Boundary=true, Self-Distillation=false, Pre-training=LRS2&3+AVS2022.09 | 92.9 | |
| Training Strategy + CM-AuxTemporal Model=DC-TCN, Video Hours=1459, Word Boundary (WB)=true2024.06 | 92.9 | |
| DC-TCNWord Boundary=true, Self-Distillation=false, Pre-training=LiRA(LRS3)2022.09 | 92.3 | |
| Training Strategy + LIRATemporal Model=DC-TCN, Video Hours=595, Word Boundary (WB)=true2024.06 | 92.3 | |
| DC-TCN (Ensemble)Word Boundary=false, Pre-training=LRS2&3+AVS, Ensemble=true2022.09 | 92.1 | |
| DC-TCNWord Boundary=true, Self-Distillation=false, Pre-training=Scratch2022.09 | 92.1 | |
| Training StrategyTemporal Model=DC-TCN, Video Hours=157, Word Boundary (WB)=true2024.06 | 92.1 | |
| MTLAMTemporal Model=DC-TCN, Video Hours=157, Word Boundary (WB)=false2024.06 | 91.7 | |
| DC-TCN (Ensemble)Word Boundary=false, Pre-training=Scratch, Ensemble=true2022.09 | 91.6 | |
| DC-TCNWord Boundary=false, Self-Distillation=false, Pre-training=LRS2&3+AVS2022.09 | 91.1 | |
| DC-TCNWord Boundary=false, Self-Distillation=false, Pre-training=Scratch2022.09 | 90.4 | |
| Training StrategyTemporal Model=DC-TCN, Video Hours=157, Word Boundary (WB)=false2024.06 | 90.4 | |
| Koumparoulis et al.Temporal Model=Transformer, Video Hours=157, Word Boundary (WB)=false2024.06 | 89.5 | |
| NetVLADTemporal Model=MS-TCN, Video Hours=157, Word Boundary (WB)=false2024.06 | 89.4 | |
| ResNet-18 + BLSTMWord Boundary=true2022.09 | 88.8 | |
| ResNet-18 + MS-TCN + S.D.Word Boundary=false, Self-Distillation=true2022.09 | 88.5 | |
| MVMTemporal Model=MS-TCN, Video Hours=157, Word Boundary (WB)=false2024.06 | 88.5 | |
| ResNet-18 + DC-TCNWord Boundary=false2022.09 | 88.4 | |
| ResNet-18 + BGRUWord Boundary=true2022.09 | 88.4 | |
| Ma et al.Temporal Model=DC-TCN, Video Hours=157, Word Boundary (WB)=false2024.06 | 88.4 | |
| Feng et al.Temporal Model=BiGRU, Video Hours=157, Word Boundary (WB)=true2024.06 | 88.4 | |
| WPCL + APFFTemporal Model=MS-TCN, Video Hours=157, Word Boundary (WB)=false2024.06 | 88.3 | |
| LIRATemporal Model=Conformer, Video Hours=590, Word Boundary (WB)=false2024.06 | 88.1 | |
| Born-AgainTemporal Model=MS-TCN, Video Hours=157, Word Boundary (WB)=false2024.06 | 87.9 | |
| Feng et al.Temporal Model=BiGRU, Video Hours=157, Word Boundary (WB)=false2024.06 | 86.2 | |
| ResNet-18 + MS-TCNWord Boundary=false2022.09 | 85.3 | |
| ResNet-18 + BGRU + CutoutWord Boundary=false2022.09 | 85 | |
| ResNet-18 + BGRUWord Boundary=false2022.09 | 85 | |
| ResNet-18 + BLSTMWord Boundary=false2022.09 | 84.3 | |
| 2*3D-CNN + BLSTMWord Boundary=false2022.09 | 84.1 | |
| ResNet-34 + BLSTMWord Boundary=false2022.09 | 83 | |
| 3D-CNNWord Boundary=false2022.09 | 61.1 |