Audiovisual Masked Autoencoders

Benchmark Model Rank Results
audio-classification-on-epic-kitchens-100Audiovisual Masked Autoencoder (Audiovisual, Single)#1Top-1 Action: 46.0Top-1 Noun: 56.4Top-1 Verb: 71.4
audio-classification-on-epic-kitchens-100Audiovisual Masked Autoencoder (Video-only, Single)#2Top-1 Action: 45.8Top-1 Noun: 55.9Top-1 Verb: 70.8
audio-classification-on-epic-kitchens-100Audiovisual Masked Autoencoder (Audio-only, Single)#3Top-1 Action: 19.7Top-1 Noun: 27.2Top-1 Verb: 52.7
audio-classification-on-audiosetAudiovisual Masked Autoencoder (Audiovisual, Single)#2Test mAP: 0.518
audio-classification-on-audiosetAudiovisual Masked Autoencoder (Audio-only, Single)#32Test mAP: 0.466
audio-classification-on-vggsoundAudiovisual Masked Autoencoder (Audiovisual, Single)#6Top 1 Accuracy: 65.0
audio-classification-on-vggsoundAudiovisual Masked Autoencoder (Audio-only, Single)#10Top 1 Accuracy: 57.2