PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning.
Lin XuYilin ZhaoDaquan ZhouZhijie LinSee-Kiong NgJiashi FengPublished in: CoRR (2024)
Keyphrases
- parameter free
- video images
- video frames
- video sequences
- natural language descriptions
- low frame rate
- static images
- video content
- successive frames
- fully automatic
- video dataset
- video data
- digital photos
- input image
- input video
- video database
- visual data
- image retrieval
- video analysis
- moving camera
- high frame rate
- key frames
- video signals
- data mining
- feature selection
- video collections
- multimedia
- object recognition
- textual descriptions
- video surveillance
- dynamic textures
- video clips
- sports video
- action recognition
- databases
- space time
- video segments
- online video
- similarity measure
- video shots
- video event
- computer vision