Learn More
In this paper, we present a motion-adaptive algorithm for streaming object segmentation in monocular video sequences. To segment each frame, we fuse the cues from color, spatiotemporal contrast, and the bilayer labels on the previous frame in a two-frame graph. In the graph the coherence factor, the weight of the temporal smoothing term, is online estimated(More)
Object pose estimation in bin-picking task is an important problem in the industry. Since objects shapes are very different, there not exists a general method now to deal with all kinds of objects. Here we use convex hulls of the planes of the objects to do object 3D pose estimation. This method ignores details of objects shapes, and is robust in matching.(More)
  • 1