JSON

Recently, we have been trying to find ways to let AI really understand our "director's intent", so that the video images generated by AI can meet our expectations. From simple text descriptions to increasingly complex parameterization, we are getting closer and closer to the goal of precise control of the image. Traditional "one-sentence" cues, such as "a girl walking in the rain", often bring great randomness to AI-generated videos. The girl's dress, her mood, the size of the rain, the way the camera moves ...... these key details are all left to the AI to "guess". This may be fun when looking for inspiration, but when it comes to precise execution...