MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video Recognition

by   Chao-Yuan Wu, et al.

While today's video recognition systems parse snapshots or short clips accurately, they cannot connect the dots and reason across a longer range of time yet. Most existing video architectures can only process <5 seconds of a video without hitting the computation or memory bottlenecks. In this paper, we propose a new strategy to overcome this challenge. Instead of trying to process more frames at once like most existing methods, we propose to process videos in an online fashion and cache "memory" at each iteration. Through the memory, the model can reference prior context for long-term modeling, with only a marginal cost. Based on this idea, we build MeMViT, a Memory-augmented Multiscale Vision Transformer, that has a temporal support 30x longer than existing models with only 4.5 need >3,000 increased temporal support enabled by MeMViT brings large gains in recognition accuracy consistently. MeMViT obtains state-of-the-art results on the AVA, EPIC-Kitchens-100 action classification, and action anticipation datasets. Code and models will be made publicly available.



There are no comments yet.


page 1

page 2

page 3

page 4


TALLFormer: Temporal Action Localization with Long-memory Transformer

Most modern approaches in temporal action localization divide this probl...

Long-Term Feature Banks for Detailed Video Understanding

To understand the world, we humans constantly need to relate the present...

Towards Long-Form Video Understanding

Our world offers a never-ending stream of visual stimuli, yet today's vi...

ECO: Efficient Convolutional Network for Online Video Understanding

The state of the art in video understanding suffers from two problems: (...

Beyond Short Clips: End-to-End Video-Level Learning with Collaborative Memories

The standard way of training video models entails sampling at each itera...

An Annotated Video Dataset for Computing Video Memorability

Using a collection of publicly available links to short form video clips...

Memory Warps for Learning Long-Term Online Video Representations

This paper proposes a novel memory-based online video representation tha...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.