The invention discloses an internet video author identity recognition method based on multi-feature fusion. The method includes the steps; inputting a video, and performing uniform down-sampling on video frame images; extracting background camera motion vectors from any one extracted frame image and a preorder frame image thereof, and calculating camera motion characteristics of the video images; calculating average shot length and mutant shot ratio based on video shot segmentation to obtain video structure characteristics; fusing the characteristics, and learning and recognizing whether an identity of a video producer is a professional video producer or an amateur video producer by the aid of a support vector machine classifier. According to the method, professional video and amateur video production characteristics are fully considered, camera motion characteristics and structure characteristics in internet videos are fused, and the support vector machine classifier can accurately learn and recognize whether the identity of the video producer is the professional video producer or the amateur video producer.