HumanML3D
A multimodal dataset in the field of human motion generation and understanding, containing text descriptions paired with corresponding human motion sequences. It is commonly used for evaluating tasks such as motion generation and retrieval.
Timeline (last 90 days)
Motion VAE is a class of methods that applies variational autoencoders to human motion sequence modeling; its encoder typically uses a Transformer or graph convolutional network, the decoder outputs a skeletal pose sequence, and the training objective is to maximize the ELBO.
The study constructed controlled multi-schema motion VAEs on the HumanML3D dataset, with the models sharing the backbone and decoder structure.
All Facts (2)
Motion VAE is a class of methods that applies variational autoencoders to human motion sequence modeling; its encoder typically uses a Transformer or graph convolutional network, the decoder outputs a skeletal pose sequence, and the training objective is to maximize the ELBO.
50%UnverifiedThe study constructed controlled multi-schema motion VAEs on the HumanML3D dataset, with the models sharing the backbone and decoder structure.
50%