Active Speaker Localization on EasyCom (test)
86.32mAPOurs AV(cor+eng+box)
Evaluation Results
| Method | Links | |
|---|---|---|
| Ours AV(cor+eng+box)Audio Channels=4, Modality=Audio-Visual, Features=Cross-correlation, Energy, Bounding box2022.01 | 86.32 | |
| Ours AV(cor+box)Audio Channels=4, Modality=Audio-Visual, Features=Cross-correlation, Bounding box2022.01 | 86.25 | |
| Ours AV(spec)Audio Channels=4, Modality=Audio-Visual, Features=Spectrogram2022.01 | 85.49 | |
| Ours AV(cor)Audio Channels=4, Modality=Audio-Visual, Features=Cross-correlation2022.01 | 84.14 | |
| Ours AV(cor+eng)Audio Channels=4, Modality=Audio-Visual, Features=Cross-correlation, Energy2022.01 | 83.32 | |
| Ours AV(spec)-2chAudio Channels=2, Modality=Audio-Visual, Features=Spectrogram2022.01 | 83.3 | |
| Ours AV(cor)-2chAudio Channels=2, Modality=Audio-Visual, Features=Cross-correlation2022.01 | 80 | |
| AV-rawaudioAudio Channels=4, Modality=Audio-Visual2022.01 | 72.32 | |
| TalkNet (AVA)Training Data=AVA, Audio Channels=1, Modality=Audio-Visual2022.01 | 69.13 | |
| MRC (EasyCom)Training Data=EasyCom, Modality=Video-only2022.01 | 64.24 | |
| Ours AV(eng)Audio Channels=4, Modality=Audio-Visual, Features=Energy2022.01 | 62.68 | |
| BinauralAVLocAudio Channels=2, Modality=Audio-Visual2022.01 | 60.75 | |
| DOA+imageAudio Channels=4, Modality=Audio-Visual2022.01 | 54.27 | |
| DOA+headboxAudio Channels=4, Modality=Audio-Visual2022.01 | 52.62 | |
| MRC (AVA)Training Data=AVA, Modality=Video-only2022.01 | 46.6 | |
| TalkNet (EasyCom)Training Data=EasyCom, Audio Channels=1, Modality=Audio-Visual2022.01 | 44.24 |