Temporal U-Nets for Video Summarization with Scene and Action Recognition.
Heeseung KwonWoohyun ShimMinsu ChoPublished in: ICCV Workshops (2019)
Keyphrases
- action recognition
- video summarization
- video sequences
- surveillance videos
- human actions
- temporal information
- audio visual
- bag of words
- video content
- atomic actions
- activity recognition
- computer vision
- d scene
- event detection
- human detection
- video data
- space time
- visual data
- key frames
- human activities
- spatio temporal
- image sequences
- dynamic scenes
- video retrieval
- three dimensional
- surveillance system
- low level features
- multi camera
- moving camera
- video frames
- multi modal
- mid level
- moving objects
- video scene
- object recognition
- video surveillance
- visual words
- object classification
- motion vectors