Long-Term Feature Banks for Detailed Video Understanding

December 12, 2018 · Entered Twilight · 🏛 Computer Vision and Pattern Recognition

"Last commit was 5.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: CODE_OF_CONDUCT.md, CONTRIBUTING.md, DATASET.md, GETTING_STARTED.md, INSTALL.md, LICENSE, README.md, caffe2_customized_ops, configs, dataset_tools, figs, lib, tools

Authors Chao-Yuan Wu, Christoph Feichtenhofer, Haoqi Fan, Kaiming He, Philipp Krähenbühl, Ross Girshick arXiv ID 1812.05038 Category cs.CV: Computer Vision Citations 505 Venue Computer Vision and Pattern Recognition Repository https://github.com/facebookresearch/video-long-term-feature-banks ⭐ 384 Last Checked 2 months ago

Abstract

To understand the world, we humans constantly need to relate the present to the past, and put events in context. In this paper, we enable existing video models to do the same. We propose a long-term feature bank---supportive information extracted over the entire span of a video---to augment state-of-the-art video models that otherwise would only view short clips of 2-5 seconds. Our experiments demonstrate that augmenting 3D convolutional networks with a long-term feature bank yields state-of-the-art results on three challenging video datasets: AVA, EPIC-Kitchens, and Charades.