Curriculum Learning for Recurrent Video Object Segmentation

Maria Gonzalez-i-Calabuig
Maria Gonzalez-i-Calabuig
Xavier Giró-i-Nieto
Xavier Giró-i-Nieto
Cited by: 0|Bibtex|Views3|Links

Abstract:

Video object segmentation can be understood as a sequence-to-sequence task that can benefit from the curriculum learning strategies for better and faster training of deep neural networks. This work explores different schedule sampling and frame skipping variations to significantly improve the performance of a recurrent architecture. Our...More

Code:

Data:

Your rating :
0

 

Tags
Comments