Here, the Maximum Entropy model is implemented for multi-modal video analysis to annotate video semantics.

• 本文用最大熵模型对视频的多模态特征进行建模，对视频进行语义标注。