MCG-NJU/videomae-base is a VideoMAE (Masked Autoencoders for Video) base model pre-trained in a self-supervised manner on video data. It can be used for video classification and as a strong initialization for fine-tuning on various video understanding tasks. The model is provided via Hugging Face and is popular among computer vision researchers working on video models.
In the Other AI space, Videomae Base takes a focused approach. It focuses on learning strong video representations from unlabeled video data for downstream video understanding tasks. Videomae Base is an open-source project aimed at machine learning researchers and video AI developers. The project is open source (Open Source). It runs on the web and API.
It is developed by Nanjing University (China), and the product first shipped in 2022. The project is developed in the open on GitHub with 1.8k stars. Among its 3 catalogued features are Video Classification, pre-training, and transformers. It exposes integrations via a public API.
Latest indexed changes and source events
MCG-NJU/videomae-base verified by the PulseGate indexer
Other apps tracked under the same category.
