A thousand frames in just a few words: Lingual description of videos through latent topics and sparse object stitching

Keyphrases

• Natural Language 100%
• Image Description 100%
• Sparse Object 100%
• Latent Topics 100%
• Human Description 100%
• Computer Vision 50%
• Long-form 50%
• Language Model 50%
• Relevant Content 50%
• Object-oriented Languages 50%
• Hybrid System 50%
• Nearest Neighbor 50%
• Object Detection Algorithm 50%
• Natural Language Description 50%
• Middle Level 50%
• Training Image 50%
• Video Description 50%
• Latent Topic Models 50%
• Keyword Annotation 50%
• Concept Detector 50%
• Probabilistic Nearest Neighbor 50%
• System Output 50%
• Module Level 50%

Computer Science

• Top-Down Approach 100%
• Annotation 50%
• Language Modeling 50%
• Object Detection 50%
• Training Image 50%
• Concept Detector 50%
• Computer Vision 50%