Intelligent video compression method and system based on video technology

The invention discloses an intelligent video compression method and system based on a video technology. The method comprises the following steps: dividing an input original video frame into a key frame and a non-key frame; encoding the key frame into a low-dimensional implicit vector, and decoding the key frame into a reconstructed key frame; estimating a motion vector between the non-key frame and an adjacent key frame, and generating a predicted non-key frame; encoding the residual error between the non-key frame and the predicted non-key frame into a low-dimensional residual error implicit vector, and decoding the residual error into a reconstructed residual error; adding the reconstructed residual error and the predicted non-key frame to obtain a reconstructed non-key frame; judging the true and false degree between the reconstructed video frame and the original video frame; measuring a pixel level difference between the reconstructed video frame and the original video frame; and optimizing and balancing