Rethinking Video ViTs: Sparse Video Tubes for Joint Image and Video Learning

12/06/2022
by   AJ Piergiovanni, et al.
0

We present a simple approach which can turn a ViT encoder into an efficient video model, which can seamlessly work with both image and video inputs. By sparsely sampling the inputs, the model is able to do training and inference from both inputs. The model is easily scalable and can be adapted to large-scale pre-trained ViTs without requiring full finetuning. The model achieves SOTA results and the code will be open-sourced.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset