Abstract | ||
---|---|---|
Extracting background features for estimating the camera path is a key step in many video editing and enhancement applications. Existing approaches often fail on highly dynamic videos that are shot by moving cameras and contain severe foreground occlusion. Based on existing theories, we present a new, practical method that can reliably identify background features in complex video, leading to accurate camera path estimation and background layering. Our approach contains a local motion analysis step and a global optimization step. We first divide the input video into overlapping temporal windows, and extract local motion clusters in each window. We form a directed graph from these local clusters, and identify background ones by finding a minimal path through the graph using optimization. We show that our method significantly outperforms other alternatives, and can be directly used to improve common video editing applications such as stabilization, compositing and background reconstruction. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1145/2980179.2980243 | ACM Trans. Graph. |
Keywords | Field | DocType |
Feature point trajectory,background detection,video enhancement,video stabilization,camera path estimation | Graph,Computer vision,Computer graphics (images),Global optimization,Computer science,Image stabilization,Directed graph,Video tracking,Video editing,Artificial intelligence,Motion analysis,Compositing | Journal |
Volume | Issue | ISSN |
35 | 6 | 0730-0301 |
Citations | PageRank | References |
9 | 0.49 | 53 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Fang-Lue Zhang | 1 | 269 | 15.60 |
Xian Wu | 2 | 18 | 3.00 |
Hao-Tian Zhang | 3 | 23 | 2.43 |
Jue Wang | 4 | 2871 | 155.89 |
Shi-Min Hu | 5 | 3466 | 188.22 |