nateraw / video-llava

Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

• Public
• 306.3K runs
• GitHub
• Paper
• License