Human activity recognition (HAR) has multifaceted applications due to its worldly usage of
acquisition devices such as smartphones, video cameras, and its ability to capture human …

Understanding actions in videos remains a significant challenge in computer vision, which
has been the subject of several pieces of research in the last decades. Convolutional neural …

Pre-training video transformers on extra large-scale datasets is generally required to
achieve premier performance on relatively small datasets. In this paper, we show that video …

In this paper, we study Multiscale Vision Transformers (MViTv2) as a unified architecture for
image and video classification, as well as object detection. We present an improved version …

Capitalizing on large pre-trained models for various downstream tasks of interest have
recently emerged with promising performance. Due to the ever-growing model size, the …

Video recognition has been dominated by the end-to-end learning paradigm–first initializing
a video recognition model with weights of a pretrained image model and then conducting …

It is a challenging task to learn discriminative representation from images and videos, due to
large local redundancy and complex global dependency in these visual data. Convolution …

Video understanding requires reasoning at multiple spatiotemporal resolutions--from short
fine-grained motions to events taking place over longer durations. Although transformer …

The vision community is witnessing a modeling shift from CNNs to Transformers, where pure
Transformer architectures have attained top accuracy on the major video recognition …

Abstract We present Multiscale Vision Transformers (MViT) for video and image recognition,
by connecting the seminal idea of multiscale feature hierarchies with transformer models …