Emotional Mimicry Intensity Estimation on Hume-Vidmimic2 (val)
0.5851Average Pearson Correlation (p-bar)EmoViT+HuBERT+ChatGLM3
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| EmoViT+HuBERT+ChatGLM3Modality=Audio+Video, Features=-2026.03 | 0.5851 | 0.7155 | 0.6159 | 0.6303 | 0.3488 | 0.6174 | 0.5793 | |
| ChatGLM3 + WavLM-Large + ViTModality=Text + Audio + Face, Fusion Strategy=concat2026.03 | 0.4786 | 0.5242 | 0.4709 | 0.4415 | 0.5846 | 0.4531 | 0.3971 | |
| ChatGLM3 + WavLM-Large + ViTModality=Text + Audio + Face, Fusion Strategy=blending2026.03 | 0.4673 | 0.5125 | 0.463 | 0.4338 | 0.5599 | 0.435 | 0.3995 | |
| GigaChat + MT-EmotiMobileViT + HuBERTModality=Text + Video + Audio, Fusion=blending2026.03 | 0.446 | 0.5197 | 0.4491 | 0.3916 | 0.4981 | 0.4192 | 0.3983 | |
| OpenAI + MT-EmotiMobileViT + HuBERTModality=Text + Video + Audio, Fusion=blending2026.03 | 0.4451 | 0.5203 | 0.4461 | 0.3938 | 0.5054 | 0.4015 | 0.4036 | |
| GigaChat + MT-EmotiMobileFaceNet + HuBERTModality=Text + Video + Audio, Fusion=blending2026.03 | 0.4338 | 0.4955 | 0.4332 | 0.3923 | 0.464 | 0.4282 | 0.3863 | |
| GigaChat + MT-EmotiMobileFaceNetModality=Text + Video, Fusion=blending2026.03 | 0.4231 | 0.4897 | 0.4107 | 0.3806 | 0.4824 | 0.4131 | 0.3618 | |
| OpenAI + MT-EmotiMobileFaceNet + HuBERTModality=Text + Video + Audio, Fusion=blending2026.03 | 0.4225 | 0.5073 | 0.4044 | 0.3696 | 0.4801 | 0.3822 | 0.3912 | |
| RoBERTa + MT-EmotiMobileFaceNet + HuBERTModality=Text + Video + Audio, Fusion=blending2026.03 | 0.4223 | 0.462 | 0.4095 | 0.3848 | 0.4936 | 0.3876 | 0.3965 | |
| GigaChat + MT-EmotiMobileFaceNet + HuBERTModality=Text + Video + Audio, Fusion=early2026.03 | 0.4204 | 0.4823 | 0.4013 | 0.4267 | 0.4496 | 0.3765 | 0.3983 | |
| GigaChat + MT-EmotiMobileFaceNetModality=Text + Video, Fusion=early2026.03 | 0.4199 | 0.4906 | 0.3988 | 0.4121 | 0.442 | 0.3876 | 0.393 | |
| RoBERTa + MT-EmotiMobileViT + HuBERTModality=Text + Video + Audio, Fusion=blending2026.03 | 0.4192 | 0.4603 | 0.4104 | 0.3844 | 0.4935 | 0.3802 | 0.3864 | |
| OpenAI + HuBERTModality=Text + Audio, Fusion=blending2026.03 | 0.4125 | 0.5114 | 0.4006 | 0.3776 | 0.4863 | 0.3645 | 0.3344 | |
| GigaChat + HuBERTModality=Text + Audio, Fusion=blending2026.03 | 0.4106 | 0.4955 | 0.3912 | 0.3881 | 0.4619 | 0.3922 | 0.3348 | |
| OpenAI + MT-EmotiMobileViTModality=Text + Video, Fusion=blending2026.03 | 0.4103 | 0.4932 | 0.409 | 0.3327 | 0.4571 | 0.3771 | 0.393 | |
| GigaChat + MT-EmotiMobileViTModality=Text + Video, Fusion=blending2026.03 | 0.4103 | 0.4932 | 0.409 | 0.3327 | 0.4571 | 0.3711 | 0.393 | |
| RoBERTa + MT-EmotiMobileFaceNetModality=Text + Video, Fusion=blending2026.03 | 0.4074 | 0.4408 | 0.3877 | 0.3632 | 0.5064 | 0.365 | 0.3814 | |
| RoBERTa + MT-EmotiMobileViTModality=Text + Video, Fusion=blending2026.03 | 0.4028 | 0.4379 | 0.3911 | 0.358 | 0.4802 | 0.3656 | 0.384 | |
| OpenAI (small)Modality=Text, Features=Embeddings (mean)2026.03 | 0.4011 | 0.5113 | 0.3825 | 0.3687 | 0.4847 | 0.3363 | 0.3233 | |
| GigaChat + HuBERTModality=Text + Audio, Fusion=early2026.03 | 0.4011 | 0.4794 | 0.3733 | 0.4182 | 0.4449 | 0.3516 | 0.3395 | |
| GigaChatModality=Text, Features=Embeddings (mean)2026.03 | 0.4001 | 0.4888 | 0.3648 | 0.3732 | 0.4862 | 0.3726 | 0.3151 | |
| RoBERTa + HuBERTModality=Text + Audio, Fusion=blending2026.03 | 0.3974 | 0.4644 | 0.3727 | 0.3814 | 0.4755 | 0.3477 | 0.343 | |
| Wav2Vec2.0+VADModality=Audio, Features=-2026.03 | 0.389 | — | — | — | — | — | — | |
| OpenAI + MT-EmotiMobileFaceNetModality=Text + Video, Fusion=blending2026.03 | 0.3887 | 0.4779 | 0.3593 | 0.3369 | 0.4373 | 0.3452 | 0.3758 | |
| SentenceE5Modality=Text, Features=Embeddings (mean)2026.03 | 0.3777 | 0.4719 | 0.3497 | 0.3417 | 0.4553 | 0.3293 | 0.3184 | |
| RoBERTa (emotional)Modality=Text, Features=Embeddings (mean)2026.03 | 0.3763 | 0.4558 | 0.3328 | 0.3427 | 0.4964 | 0.3201 | 0.3099 | |
| SentenceBERTModality=Text, Features=Embeddings (mean)2026.03 | 0.3756 | 0.471 | 0.3585 | 0.3589 | 0.4347 | 0.3186 | 0.3119 | |
| RoBERTa (emotional)Modality=Text, Features=Embeddings (STAT)2026.03 | 0.3697 | 0.4501 | 0.3326 | 0.3173 | 0.4843 | 0.3331 | 0.3008 | |
| Sentence all-MiniLMModality=Text, Features=Embeddings (mean)2026.03 | 0.3537 | 0.4552 | 0.3076 | 0.3371 | 0.4111 | 0.3093 | 0.3023 | |
| ResNet18+AUs+Wav2Vec2.0Modality=Audio+Video, Features=-2026.03 | 0.3288 | — | — | — | — | — | — | |
| HuBERTModality=Audio, Features=Embeddings (mean)2026.03 | 0.3045 | 0.3644 | 0.3179 | 0.2987 | 0.3111 | 0.2848 | 0.2499 | |
| wav2vec 2.0 + MT-EmotiMobileFaceNetModality=Audio + Video2026.03 | 0.2898 | 0.3041 | 0.3004 | 0.2584 | 0.3148 | 0.316 | 0.2452 | |
| wav2vec 2.0 + MT-EmotiMobileViTModality=Audio + Video2026.03 | 0.2829 | 0.3011 | 0.2968 | 0.2595 | 0.3074 | 0.3171 | 0.2152 | |
| HuBERTModality=Audio, Features=Embeddings (STAT)2026.03 | 0.2729 | 0.3585 | 0.2454 | 0.2432 | 0.3065 | 0.2566 | 0.2275 | |
| ViT+wav2Vec2Modality=Audio+Video, Features=-2026.03 | 0.25 | — | — | — | — | — | — | |
| wav2Vec2Modality=Audio, Features=-2026.03 | 0.24 | — | — | — | — | — | — | |
| wav2vec 2.0Modality=Audio, Features=Embeddings (STAT)2026.03 | 0.2311 | 0.3006 | 0.1659 | 0.2559 | 0.3198 | 0.1844 | 0.1602 | |
| MT-EmotiMobileViTModality=Faces, Features=Scores (STAT)2026.03 | 0.1776 | 0.0619 | 0.2623 | 0.1247 | 0.0773 | 0.2544 | 0.2849 | |
| MT-EmotiMobileFaceNetModality=Faces, Features=Scores (STAT)2026.03 | 0.1732 | 0.0285 | 0.2498 | 0.1318 | 0.097 | 0.2543 | 0.2776 | |
| MT-EmotiMobileViTModality=Faces, Features=Embeddings (STAT)2026.03 | 0.1683 | 0.0433 | 0.2459 | 0.1347 | 0.0779 | 0.2382 | 0.2699 | |
| MT-EmotiMobileFaceNetModality=Faces, Features=Scores (mean)2026.03 | 0.1667 | 0.0276 | 0.2367 | 0.1336 | 0.0807 | 0.2516 | 0.2699 | |
| MT-EmotiMobileFaceNetModality=Faces, Features=Embeddings (STAT)2026.03 | 0.1646 | 0.0557 | 0.238 | 0.1303 | 0.0703 | 0.2325 | 0.2605 | |
| MT-EmotiMobileViTModality=Faces, Features=Embeddings (mean)2026.03 | 0.1644 | 0.0379 | 0.2314 | 0.1387 | 0.0781 | 0.2334 | 0.2672 | |
| MT-EmotiMobileViTModality=Faces, Features=Scores (mean)2026.03 | 0.1642 | 0.0321 | 0.2484 | 0.149 | 0.0674 | 0.2399 | 0.2481 | |
| MT-EmotiMobileFaceNetModality=Faces, Features=Embeddings (mean)2026.03 | 0.1518 | 0.0215 | 0.2288 | 0.114 | 0.0692 | 0.2299 | 0.2476 | |
| wav2vec 2.0Modality=Audio, Features=Embeddings (mean)2026.03 | 0.1514 | 0.2153 | 0.1176 | 0.1834 | 0.1426 | 0.1275 | 0.1219 | |
| Baseline ViTModality=Faces, Features=-2026.03 | 0.09 | — | — | — | — | — | — |