作者: Tomer Amiaz , Sándor Fazekas , Dmitry Chetverikov , Nahum Kiryati
DOI: 10.1007/978-3-540-72823-8_73
关键词: Motion estimation 、 Level set 、 Object (computer science) 、 Optical flow 、 Image texture 、 Mathematics 、 Image (mathematics) 、 Computer vision 、 Artificial intelligence 、 Brightness 、 Segmentation
摘要: Motion estimation is usually based on the brightness constancy assumption. This assumption holds well for rigid objects with a Lambertian surface, but it less appropriate fluid and gaseous materials. For these materials variant of this assumption, which we call conservation should be employed. Under an object's can diffuse to its neighborhood. We propose method detecting regions dynamic texture in image sequences. Segmentation into static achieved by using level set scheme. The function separates images areas obeying those obey conservation. Experimental results challenging sequences demonstrate success segmentation scheme validate model.