Deep Feature Flow for Video Recognition
Authors:
- Xizhou ZhuXZXizhou Zhu
- Yuwen XiongYXYuwen Xiong
- Jifeng DaiJDJifeng Dai
- Lu YuanLYLu Yuan
- Yichen WeiYWYichen Wei
Top Cited Papers
- 1 July 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636919,p. 4141-4150
Abstract
Deep convolutional neutral networks have achieved great success on image recognition tasks. Yet, it is non-trivial to transfer the state-of-the-art image recognition networks to videos as per-frame evaluation is too slow and unaffordable. We present deep feature flow, a fast and accurate framework for video recognition. It runs the expensive convolutional sub-network only on sparse key frames and propagates their deep feature maps to other frames via a flow field. It achieves significant speedup as flow computation is relatively fast. The end-to-end training of the whole architecture significantly boosts the recognition accuracy. Deep feature flow is flexible and general. It is validated on two recent large scale video datasets. It makes a large step towards practical video recognition. Code would be released.Keywords
All Related Versions
This publication has 25 references indexed in Scilit:
- Optical Flow with Semantic Segmentation and Localized Layers2016
- Feature Space Optimization for Semantic Video Segmentation2016
- Slow and Steady Feature Analysis: Higher Order Temporal Coherence in Video2016
- Conditional Random Fields as Recurrent Neural Networks2015
- Flowing ConvNets for Human Pose Estimation in Videos2015
- EpicFlow: Edge-preserving interpolation of correspondences for optical flow2015
- DeepID-Net: Deformable deep convolutional neural networks for object detection2015
- Fully convolutional networks for semantic segmentation2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- A Unified Video Segmentation Benchmark: Annotation, Metrics and Analysis2013