Video intermediate frame determination method and device, equipment and storage medium

By performing branch feature extraction and optical flow estimation on video frames, the target intermediate frame is constructed, which solves the problems of motion blur and jitter in traditional video interpolation methods when dealing with the fast moving small objects, and achieves more efficient interpolation effect and smoothness.

CN119728899AActive Publication Date: 2025-03-28SUZHOU GAIDE PHOTOELECTRIC TECH CO LTD
11 Cites 0 Cited by

Patent Information

Application Number
CN202411883126.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-19
Publication Date
2025-03-28
Estimated Expiration
2044-12-19

AI Technical Summary

Technical Problem

Traditional video interpolation methods easily lead to motion blur and jitter when processing small objects that move quickly, and have high computational complexity, making it difficult to effectively capture the motion trajectory.

Method used

By acquiring adjacent video frames to be inserted, branch feature extraction is performed to obtain video depth features, and optical flow motion information is obtained based on optical flow estimation to construct target intermediate frames to improve interpolation effect and smoothness.

Benefits of technology

In a fast moving small object scene, the motion trajectory can be better captured, the interpolation effect can be improved, the smoothness of the interpolation result can be improved, and the computational complexity can be reduced.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a video intermediate frame determination method and device, equipment and a storage medium. The method comprises the following steps: acquiring adjacent video frames to be subjected to frame insertion; branch feature extraction is carried out on the adjacent video frames to obtain video depth features, and the branch feature extraction at least comprises coarse-grained feature extraction and fine-grained feature extraction; performing optical flow estimation on the video depth features to obtain optical flow motion information; and constructing a target intermediate frame according to the optical flow motion information and the adjacent video frames. According to the technical scheme of the invention, in a small object scene in rapid motion, the motion trail can be better captured, the smoothness of the frame insertion result of the intermediate frame of the video is ensured, the error problem during the motion of the small object is solved, and the frame insertion effect is obviously improved.
Need to check novelty before this filing date? Find Prior Art