Uncovering Hidden Connections: Iterative Tracking and Reasoning for Video-grounded Dialog.
Haoyu ZhangMeng LiuYaowei WangDa CaoWeili GuanLiqiang NiePublished in: CoRR (2023)
Keyphrases
- object segmentation and tracking
- face detection and tracking
- video images
- real time
- object detection and tracking
- video sequences
- video surveillance
- person detection
- surveillance videos
- moving camera
- video data
- visual tracking
- low frame rate
- particle filter
- temporal continuity
- multimedia
- video frames
- video streams
- image frames
- video analysis
- object tracking
- input video
- knowledge base
- video tracking
- successive frames
- object motion
- live video
- natural language
- stationary camera
- knowledge representation
- video content
- wide area motion imagery
- computer vision
- articulated human motion
- crowded scenes
- text detection
- real time face tracking
- dynamic scenes
- image sequences
- kalman filter
- video clips
- video processing
- mixed initiative
- conversational agents
- moving target
- video database
- particle filtering
- temporal consistency
- temporal information
- appearance model
- mean shift
- objects in video sequences
- space time
- video dataset