Learning Instance Motion Segmentation With Geometric Embedding

Most existing deep learning-based motion segmentation methods treat motion segmentation as a binary segmentation problem, which is generally not the real case in dynamic scenes. In addition, the object and camera motion are often mixed, making the motion segmentation problem difficult. This paper pr...

Popoln opis

Bibliografske podrobnosti

Principais autores:	Zhen Leng, Jing Chen, Songnan Lin
Format:	Artigo
Jezik:	Inglês
Izdano:	IEEE 2021-01-01
Serija:	IEEE Access
Teme:	Supervised learning motion segmentation video object segmentation
Online dostop:	https://ieeexplore.ieee.org/document/9380630/
Oznake:	Označite Brez oznak, prvi označite!